{
"metadata": {
"name": ""
},
"nbformat": 3,
"nbformat_minor": 0,
"worksheets": [
{
"cells": [
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# And the Oscar Goes to...\n",
"\n",
"\n",
"
![](\"http://hpsupporters.com/wp-content/uploads/2012/01/oscar-3.jpg\")
\n",
"
\n",
"\n",
""
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"%matplotlib inline\n",
"\n",
"import gc\n",
"import json, re\n",
"import unicodedata\n",
"import random, math\n",
"from pattern import web\n",
"from HTMLParser import HTMLParser\n",
"import requests\n",
"import pandas as pd\n",
"import numpy as np\n",
"import matplotlib.pyplot as plt\n",
"\n",
"pd.set_option('display.width', 500)\n",
"pd.set_option('display.max_columns', 30)\n",
"\n",
"# set some nicer defaults for matplotlib\n",
"from matplotlib import rcParams\n",
"\n",
"api_key = 'sdat77wpkdnuyvb9bqhz2y5v' #nicks api key for rotten tomatoes\n",
"\n",
"#these colors come from colorbrewer2.org. Each is an RGB triplet\n",
"dark2_colors = [(0.10588235294117647, 0.6196078431372549, 0.4666666666666667),\n",
" (0.8509803921568627, 0.37254901960784315, 0.00784313725490196),\n",
" (0.4588235294117647, 0.4392156862745098, 0.7019607843137254),\n",
" (0.9058823529411765, 0.1607843137254902, 0.5411764705882353),\n",
" (0.4, 0.6509803921568628, 0.11764705882352941),\n",
" (0.9019607843137255, 0.6705882352941176, 0.00784313725490196),\n",
" (0.6509803921568628, 0.4627450980392157, 0.11372549019607843),\n",
" (0.4, 0.4, 0.4)]\n",
"\n",
"rcParams['figure.figsize'] = (10, 6)\n",
"rcParams['figure.dpi'] = 150\n",
"rcParams['axes.color_cycle'] = dark2_colors\n",
"rcParams['lines.linewidth'] = 2\n",
"rcParams['axes.grid'] = False\n",
"rcParams['axes.facecolor'] = 'white'\n",
"rcParams['font.size'] = 14\n",
"rcParams['patch.edgecolor'] = 'none'\n",
"\n",
"def remove_border(axes=None, top=False, right=False, left=True, bottom=True):\n",
" \"\"\"\n",
" Minimize chartjunk by stripping out unnecesary plot borders and axis ticks\n",
" \n",
" The top/right/left/bottom keywords toggle whether the corresponding plot border is drawn\n",
" \"\"\"\n",
" ax = axes or plt.gca()\n",
" ax.spines['top'].set_visible(top)\n",
" ax.spines['right'].set_visible(right)\n",
" ax.spines['left'].set_visible(left)\n",
" ax.spines['bottom'].set_visible(bottom)\n",
" \n",
" #turn off all ticks\n",
" ax.yaxis.set_ticks_position('none')\n",
" ax.xaxis.set_ticks_position('none')\n",
" \n",
" #now re-enable visibles\n",
" if top:\n",
" ax.xaxis.tick_top()\n",
" if bottom:\n",
" ax.xaxis.tick_bottom()\n",
" if left:\n",
" ax.yaxis.tick_left()\n",
" if right:\n",
" ax.yaxis.tick_right()\n",
" \n",
"def histogram_style():\n",
" remove_border(left=False)\n",
" plt.grid(False)\n",
" plt.grid(axis='y', color='w', linestyle='-', lw=1)"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 106
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"**Getting the Oscar Data**\n",
"We are pulling the Oscar Award and Nomination data from [IMDB](http://www.imdb.com/oscars/). From this we construct a dictionary with the following format:\n",
"\n",
"Oscars: {\n",
" Year: {\n",
" Award: {\n",
" Winner: (name,id)\n",
" Nominations: [(name,id)]\n",
" }\n",
" }\n",
"}\n",
""
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"\"\"\"\n",
"EXAMPLE OF USAGE:\n",
"allDataEver = get_all_years()\n",
"allDataEver = repair_dict(oscarArray, allDataEver)\n",
"print allDataEver[2013]['Best Cinematography']\n",
"print get_by_award_all_years('Best Picture', allDataEver) \n",
"print get_by_year(1952)\n",
"\n",
"#official oscar titles\n",
"#note Best Picture instead of Best Achievement in Picture etc. \n",
"#note that not all oscars exist for every year, i.e. best black and white \n",
"\n",
"oscarArray = ['Best Picture', 'Best Actor in a Leading Role', \n",
"'Best Actress in a Leading Role', 'Best Actor in a Supporting Role', \n",
"'Best Actress in a Supporting Role', 'Best Actor in a Supporting Role', \n",
"'Best Director', 'Best Assistant Director', 'Best Writing, Screenplay Written Directly for the Screen', \n",
"'Best Writing, Screenplay Based on Material from Another Medium', 'Best Cinematography', \n",
"'Best Art Direction-Set Decoration', 'Best Costume Design', 'Best Sound', 'Best Film Editing', \n",
"'Best Sound Editing','Best Effects, Visual Effects', 'Best Makeup', 'Best Music, Original Song', \n",
"'Best Music, Original Score', 'Best Short Film, Animated', 'Best Short Film, Live Action', 'Best Documentary, Short Subjects', \n",
"'Best Documentary, Features', 'Best Foreign Language Film']\n",
"\"\"\"\n",
"\n",
"# HELPER FUNCTIONS\n",
"\n",
"from pattern.web import URL, DOM, plaintext, strip_between\n",
"from pattern.web import NODE, TEXT, COMMENT, ELEMENT, DOCUMENT\n",
"\n",
"# get rid of accents etc.\n",
"def fixWeirdChars(phrase):\n",
" HexEncode = ['E9','E1','E0','B7']\n",
" matchObj = re.search(r'&#x..', phrase)\n",
" if matchObj:\n",
" accentedHex = matchObj.group()\n",
" HexPart = accentedHex.split('x')[1]\n",
" if(HexPart in HexEncode):\n",
" phrase = re.sub((accentedHex+';'), 'e', phrase)\n",
" else:\n",
" Hexy = \"0x\" + HexPart\n",
" Num = int(Hexy, 16)\n",
" Char = chr(Num)\n",
" phrase = re.sub((accentedHex+';'), Char, phrase)\n",
"\n",
" matchAgain = re.search(r'&#x..', phrase)\n",
" return fixWeirdChars(phrase)\n",
" return phrase\n",
"\n",
"\n",
"def clean_unicode(s):\n",
"\ts = str(s)\n",
"\ts.encode('ascii','ignore')\n",
"\ts = fixWeirdChars(s)\n",
" \treturn str(s)\n",
"\n",
"# repair dictionary names\n",
"def repair_dict(dictionary):\n",
"\tfor year in dictionary:\n",
"\t\tfor award in dictionary[year]:\n",
"\t\t\tif(award == 'Best Motion Picture of the Year'):\n",
"\t\t\t\tdictionary[year]['Best Picture'] = dictionary[year]['Best Motion Picture of the Year']\n",
"\t\t\t\tdel dictionary[year]['Best Motion Picture of the Year']\n",
"\n",
"\t\t\tif(award == 'Best Performance by an Actor in a Leading Role'):\n",
"\t\t\t\tdictionary[year]['Best Actor in a Leading Role'] = dictionary[year]['Best Performance by an Actor in a Leading Role']\n",
"\t\t\t\tdel dictionary[year]['Best Performance by an Actor in a Leading Role']\n",
"\n",
"\t\t\tif(award == 'Best Performance by an Actress in a Leading Role'):\n",
"\t\t\t\tdictionary[year]['Best Actress in a Leading Role'] = dictionary[year]['Best Performance by an Actress in a Leading Role']\n",
"\t\t\t\tdel dictionary[year]['Best Performance by an Actress in a Leading Role']\n",
"\n",
"\t\t\tif(award == 'Best Performance by an Actress in a Supporting Role'):\n",
"\t\t\t\tdictionary[year]['Best Actress in a Supporting Role'] = dictionary[year]['Best Performance by an Actress in a Supporting Role']\n",
"\t\t\t\tdel dictionary[year]['Best Performance by an Actress in a Supporting Role']\n",
"\n",
"\t\t\tif(award == 'Best Performance by an Actor in a Supporting Role'):\n",
"\t\t\t\tdictionary[year]['Best Actor in a Supporting Role'] = dictionary[year]['Best Performance by an Actor in a Supporting Role']\n",
"\t\t\t\tdel dictionary[year]['Best Performance by an Actor in a Supporting Role']\n",
"\n",
"\t\t\tif(award == 'Best Writing, Original Screenplay'):\n",
"\t\t\t\tdictionary[year]['Best Writing, Screenplay Written Directly for the Screen'] = dictionary[year]['Best Writing, Original Screenplay']\n",
"\t\t\t\tdel dictionary[year]['Best Writing, Original Screenplay']\n",
"\n",
"\t\t\tif(award == 'Best Writing, Adapted Screenplay'):\n",
"\t\t\t\tdictionary[year]['Best Writing, Screenplay Based on Material from Another Medium'] = dictionary[year]['Best Writing, Adapted Screenplay']\n",
"\t\t\t\tdel dictionary[year]['Best Writing, Adapted Screenplay']\n",
"\n",
"\t\t\tif(award == 'Best Foreign Language Film of the Year'):\n",
"\t\t\t\tnewAwardName = award\n",
"\t\t\t\tnewAwardName = re.sub(' of the Year', '', newAwardName)\n",
"\t\t\t\tdictionary[year][newAwardName] = dictionary[year][award]\n",
"\t\t\t\tdel dictionary[year][award]\n",
"\n",
"\t\t\t# check to see if it includes best achievement\n",
"\t\t\tif('Best Achievement' in award):\n",
"\t\t\t\tif(award == 'Best Achievement in Directing'):\n",
"\t\t\t\t\tdictionary[year]['Best Director'] = dictionary[year]['Best Achievement in Directing']\n",
"\t\t\t\t\tdel dictionary[year]['Best Achievement in Directing']\n",
"\t\t\t\telse:\n",
"\t\t\t\t\tnewAwardName = award\n",
"\t\t\t\t\tif('Mixing' in award):\n",
"\t\t\t\t\t\tnewAwardName = re.sub(' Mixing', '', newAwardName)\n",
"\n",
"\t\t\t\t\tif(' and Hairstyling' in award):\n",
"\t\t\t\t\t\tnewAwardName = re.sub(' and Hairstyling', '', newAwardName)\n",
"\n",
"\t\t\t\t\tif('Music Written for Motion Pictures, Original Song' or 'Music Written for Motion Pictures, Original Score' in award):\n",
"\t\t\t\t\t\tnewAwardName = re.sub(' Written for Motion Pictures', '', newAwardName)\n",
"\n",
"\n",
"\t\t\t\t\tnewAwardName = re.sub(' Achievement in', '', newAwardName)\n",
"\n",
"\t\t\t\t\tif 'Visual Effects' in newAwardName:\n",
"\t\t\t\t\t\t#add comma\n",
"\t\t\t\t\t\tnewAwardName = 'Best Effects, Visual Effects' \n",
"\n",
"\t\t\t\t\tdictionary[year][newAwardName] = dictionary[year][award]\n",
"\t\t\t\t\tdel dictionary[year][award]\n",
"\t\t\t\t\t\n",
"\treturn dictionary\n",
"\n",
"# END OF HELPER FUNCTIONS\n",
"\n",
"\n",
"\"\"\"\n",
"Gets the movie titles for all Oscars for a given year. i.e. get_by_year(2012)\n",
"Function\n",
"--------\n",
"get_by_year\n",
"\n",
"Input int year\n",
"\n",
"Outputs a dict of the format:\n",
"'Award':\n",
"\t'nominees': ['a', 'b', 'c', 'd'] \n",
"\t'winner': 'f'\n",
"\n",
"\"\"\"\n",
"def get_by_year(year):\n",
"\n",
" url = URL(\"http://www.imdb.com/event/ev0000003/\" + str(year))\n",
" dom = DOM(url.download(cached=True))\n",
" \n",
" dictAll = {}\n",
" \n",
" awards = dom.by_class('award')\n",
" awardTitles = awards[0].by_tag('h2')\n",
" awardList = []\n",
" for award in awardTitles:\n",
" awardList.append(award.content)\n",
"\n",
" prize = awards[0].by_tag('blockquote')\n",
" for index, title in enumerate(prize[1:25]):\n",
" winner = title.by_tag('strong')[0].by_tag('a')[0].content\n",
" winner_id = str(title.by_tag('strong')[0].by_tag('a')[0].attrs['href'][-8:-1])\n",
" \n",
" nomineeList = []\n",
" for each in title.by_tag('strong')[1::]:\n",
" name = each.by_tag('a')[0].content\n",
" id = str(each.by_tag('a')[0].attrs['href'][-8:-1])\n",
" nomineeList.append((clean_unicode(name),id))\n",
" \n",
" winnersAndNominees = {}\n",
" winnersAndNominees['winner'] = (clean_unicode(winner),winner_id)\n",
" winnersAndNominees['nominees'] = nomineeList\n",
" dictAll[awardList[index]] = winnersAndNominees\n",
" return dictAll\n",
"\n",
"\n",
"\n",
"\"\"\"\n",
"Gets the movie titles for all Oscars for all years between 1935 and 2013.\n",
"Function\n",
"--------\n",
"get_all_years()\n",
"\n",
"\n",
"Outputs a dict of dicts the format:\n",
"'Year':\n",
"\t'Award':\n",
"\t\t'nominees': ['a', 'b', 'c', 'd'] \n",
"\t\t'winner': 'f'\n",
"\"\"\"\n",
"\n",
"def get_all_years():\n",
"\thugeDict = {} #lol\n",
"\tfor year in range(2003, 2014):\n",
"\t\thugeDict[year] = get_by_year(str(year))\n",
"\treturn hugeDict\n",
"\n",
"\n",
"\n",
"'''\n",
"Gets all data for every year for a specific oscar\n",
"Function\n",
"------\n",
"get_by_award_all_years(oscar, dictionary)\n",
"\n",
"Outputs a dict of the format\n",
"'Year':\n",
"\t'nominees': ['a', 'b', 'c', 'd'] \n",
"\t'winner': 'f'\n",
" \n",
"'''\n",
"def get_by_award_all_years(oscar, dictionary):\n",
"\tyearDict = {}\n",
"\tfor year in dictionary:\n",
"\t\tif oscar in dictionary[year].keys():\n",
"\t\t\tyearDict[year] = dictionary[year][oscar]\n",
"\treturn yearDict"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 89
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"# oscars = get_all_years();\n",
"# oscars = repair_dict(oscars)\n",
"# f = open(\"data/oscars.json\",\"w+\")\n",
"# json.dump(oscars,f)\n",
"# f.close()\n",
"\n",
"# load an existing dictionary\n",
"f = open(\"data/oscars.json\",\"rb\")\n",
"oscars = json.load(f)\n",
"f.close()\n",
"print oscars"
],
"language": "python",
"metadata": {},
"outputs": [
{
"output_type": "stream",
"stream": "stdout",
"text": [
"{u'2003': {u'Best Actor in a Supporting Role': {u'nominees': [[u'Catch Me If You Can', u'0264464'], [u'Chicago', u'0299658'], [u'The Hours', u'0274558'], [u'Road to Perdition', u'0257044']], u'winner': [u'Adaptation.', u'0268126']}, u'Best Documentary, Features': {u'nominees': [[u'Daughter from Danang', u'0303281'], [u'Winged Migration', u'0301727'], [u'Prisoner of Paradise', u'0316013'], [u'Spellbound', u'0334405']], u'winner': [u'Bowling for Columbine', u'0310793']}, u'Best Writing, Screenplay Written Directly for the Screen': {u'nominees': [[u'Far from Heaven', u'0297884'], [u'Gangs of New York', u'0217505'], [u'My Big Fat Greek Wedding', u'0259446'], [u'Y Tu Mame Tambien', u'0245574']], u'winner': [u'Talk to Her', u'0287467']}, u'Best Foreign Language Film': {u'nominees': [[u'The Crime of Father Amaro', u'0313196'], [u'The Man Without a Past', u'0311519'], [u'Hero', u'0299977'], [u'Zus & zo', u'0245157']], u'winner': [u'Nowhere in Africa', u'0161860']}, u'Best Short Film, Live Action': {u'nominees': [[u'Gridlock', u'0281850'], [u'Dog', u'0303350'], [u\"I'll Wait for the Next One...\", u'0325638'], [u'Johnny Flynton', u'0329245']], u'winner': [u'Der er en yndig mand', u'0340071']}, u'Best Music, Original Song': {u'nominees': [[u'Chicago', u'0299658'], [u'Frida', u'0120679'], [u'Gangs of New York', u'0217505'], [u'The Wild Thornberrys Movie', u'0282120']], u'winner': [u'8 Mile', u'0298203']}, u'Best Music, Original Score': {u'nominees': [[u'Catch Me If You Can', u'0264464'], [u'Far from Heaven', u'0297884'], [u'The Hours', u'0274558'], [u'Road to Perdition', u'0257044']], u'winner': [u'Frida', u'0120679']}, u'Best Art Direction-Set Decoration': {u'nominees': [[u'Frida', u'0120679'], [u'Gangs of New York', u'0217505'], [u'The Lord of the Rings: The Two Towers', u'0167261'], [u'Road to Perdition', u'0257044']], u'winner': [u'Chicago', u'0299658']}, u'Best Costume Design': {u'nominees': [[u'Frida', u'0120679'], [u'Gangs of New York', u'0217505'], [u'The Hours', u'0274558'], [u'The Pianist', u'0253474']], u'winner': [u'Chicago', u'0299658']}, u'Best Short Film, Animated': {u'nominees': [[u'The Cathedral', u'0351167'], [u\"Mike's New Car\", u'0323250'], [u'Mount Head', u'0351382'], [u'Rocks', u'0330801']], u'winner': [u'The Chubbchubbs!', u'0331218']}, u'Best Makeup': {u'nominees': [[u'The Time Machine', u'0268695']], u'winner': [u'Frida', u'0120679']}, u'Best Film Editing': {u'nominees': [[u'Gangs of New York', u'0217505'], [u'The Hours', u'0274558'], [u'The Lord of the Rings: The Two Towers', u'0167261'], [u'The Pianist', u'0253474']], u'winner': [u'Chicago', u'0299658']}, u'Best Animated Feature': {u'nominees': [[u'Lilo & Stitch', u'0275847'], [u'Treasure Planet', u'0133240'], [u'Ice Age', u'0268380'], [u'Spirit: Stallion of the Cimarron', u'0166813']], u'winner': [u'Spirited Away', u'0245429']}, u'Best Writing, Screenplay Based on Material from Another Medium': {u'nominees': [[u'About a Boy', u'0276751'], [u'Adaptation.', u'0268126'], [u'Chicago', u'0299658'], [u'The Hours', u'0274558']], u'winner': [u'The Pianist', u'0253474']}, u'Best Sound Editing': {u'nominees': [[u'Minority Report', u'0181689'], [u'Road to Perdition', u'0257044']], u'winner': [u'The Lord of the Rings: The Two Towers', u'0167261']}, u'Best Actress in a Supporting Role': {u'nominees': [[u'About Schmidt', u'0257360'], [u'Adaptation.', u'0268126'], [u'Chicago', u'0299658'], [u'The Hours', u'0274558']], u'winner': [u'Chicago', u'0299658']}, u'Best Cinematography': {u'nominees': [[u'Chicago', u'0299658'], [u'Far from Heaven', u'0297884'], [u'Gangs of New York', u'0217505'], [u'The Pianist', u'0253474']], u'winner': [u'Road to Perdition', u'0257044']}, u'Best Documentary, Short Subjects': {u'nominees': [[u'The Collector of Bedford Street', u'0325193'], [u'Mighty Times: The Legacy of Rosa Parks', u'0353737'], [u\"Why Can't We Be a Family Again?\", u'0344563']], u'winner': [u'Twin Towers', u'0344503']}, u'Best Sound': {u'nominees': [[u'Gangs of New York', u'0217505'], [u'The Lord of the Rings: The Two Towers', u'0167261'], [u'Road to Perdition', u'0257044'], [u'Spider-Man', u'0145487']], u'winner': [u'Chicago', u'0299658']}, u'Best Director': {u'nominees': [[u'Chicago', u'0299658'], [u'Gangs of New York', u'0217505'], [u'Talk to Her', u'0287467'], [u'The Hours', u'0274558']], u'winner': [u'The Pianist', u'0253474']}, u'Best Actor in a Leading Role': {u'nominees': [[u'About Schmidt', u'0257360'], [u'Adaptation.', u'0268126'], [u'Gangs of New York', u'0217505'], [u'The Quiet American', u'0258068']], u'winner': [u'The Pianist', u'0253474']}, u'Best Picture': {u'nominees': [[u'Gangs of New York', u'0217505'], [u'The Hours', u'0274558'], [u'The Lord of the Rings: The Two Towers', u'0167261'], [u'The Pianist', u'0253474']], u'winner': [u'Chicago', u'0299658']}, u'Best Actress in a Leading Role': {u'nominees': [[u'Chicago', u'0299658'], [u'Far from Heaven', u'0297884'], [u'Frida', u'0120679'], [u'Unfaithful', u'0250797']], u'winner': [u'The Hours', u'0274558']}, u'Best Visual Effects': {u'nominees': [[u'Spider-Man', u'0145487'], [u'Star Wars: Episode II - Attack of the Clones', u'0121765']], u'winner': [u'The Lord of the Rings: The Two Towers', u'0167261']}}, u'2006': {u'Best Actor in a Supporting Role': {u'nominees': [[u'Brokeback Mountain', u'0388795'], [u'Cinderella Man', u'0352248'], [u'Crash', u'0375679'], [u'A History of Violence', u'0399146']], u'winner': [u'Syriana', u'0365737']}, u'Best Documentary, Features': {u'nominees': [[u\"Darwin's Nightmare\", u'0424024'], [u'Enron: The Smartest Guys in the Room', u'1016268'], [u'Murderball', u'0436613'], [u'Street Fight', u'0457496']], u'winner': [u'March of the Penguins', u'0428803']}, u'Best Writing, Screenplay Written Directly for the Screen': {u'nominees': [[u'Good Night, and Good Luck.', u'0433383'], [u'Match Point', u'0416320'], [u'The Squid and the Whale', u'0367089'], [u'Syriana', u'0365737']], u'winner': [u'Crash', u'0375679']}, u'Best Foreign Language Film': {u'nominees': [[u\"Don't Tell\", u'0443446'], [u'Joyeux Noel', u'0424205'], [u'Paradise Now', u'0445620'], [u'Sophie Scholl: The Final Days', u'0426578']], u'winner': [u'Tsotsi', u'0468565']}, u'Best Film Editing': {u'nominees': [[u'Cinderella Man', u'0352248'], [u'The Constant Gardener', u'0387131'], [u'Munich', u'0408306'], [u'Walk the Line', u'0358273']], u'winner': [u'Crash', u'0375679']}, u'Best Art Direction': {u'nominees': [[u'Good Night, and Good Luck.', u'0433383'], [u'Harry Potter and the Goblet of Fire', u'0330373'], [u'King Kong', u'0360717'], [u'Pride & Prejudice', u'0414387']], u'winner': [u'Memoirs of a Geisha', u'0397535']}, u'Best Short Film, Live Action': {u'nominees': [[u'The Runaway', u'0452139'], [u'Cashback', u'0409799'], [u'Our Time Is Up', u'0427617'], [u'The Last Farm', u'0461847']], u'winner': [u'Six Shooter', u'0425458']}, u'Best Music, Original Song': {u'nominees': [[u'Crash', u'0375679'], [u'Transamerica', u'0407265']], u'winner': [u'Hustle & Flow', u'0410097']}, u'Best Music, Original Score': {u'nominees': [[u'The Constant Gardener', u'0387131'], [u'Memoirs of a Geisha', u'0397535'], [u'Munich', u'0408306'], [u'Pride & Prejudice', u'0414387']], u'winner': [u'Brokeback Mountain', u'0388795']}, u'Best Short Film, Animated': {u'nominees': [[u'9', u'0443424'], [u'Badgered', u'0455912'], [u'The Mysterious Geographic Explorations of Jasper Morello', u'0469146'], [u'One Man Band', u'0479113']], u'winner': [u'The Moon and the Son: An Imagined Conversation', u'0449619']}, u'Best Makeup': {u'nominees': [[u'Cinderella Man', u'0352248'], [u'Star Wars: Episode III - Revenge of the Sith', u'0121766']], u'winner': [u'The Chronicles of Narnia: The Lion, the Witch and the Wardrobe', u'0363771']}, u'Best Costume Design': {u'nominees': [[u'Charlie and the Chocolate Factory', u'0367594'], [u'Mrs Henderson Presents', u'0413015'], [u'Pride & Prejudice', u'0414387'], [u'Walk the Line', u'0358273']], u'winner': [u'Memoirs of a Geisha', u'0397535']}, u'Best Writing, Screenplay Based on Material from Another Medium': {u'nominees': [[u'Capote', u'0379725'], [u'The Constant Gardener', u'0387131'], [u'A History of Violence', u'0399146'], [u'Munich', u'0408306']], u'winner': [u'Brokeback Mountain', u'0388795']}, u'Best Sound Editing': {u'nominees': [[u'Memoirs of a Geisha', u'0397535'], [u'War of the Worlds', u'0407304']], u'winner': [u'King Kong', u'0360717']}, u'Best Actress in a Supporting Role': {u'nominees': [[u'Brokeback Mountain', u'0388795'], [u'Capote', u'0379725'], [u'Junebug', u'0418773'], [u'North Country', u'0395972']], u'winner': [u'The Constant Gardener', u'0387131']}, u'Best Cinematography': {u'nominees': [[u'Batman Begins', u'0372784'], [u'Brokeback Mountain', u'0388795'], [u'Good Night, and Good Luck.', u'0433383'], [u'The New World', u'0402399']], u'winner': [u'Memoirs of a Geisha', u'0397535']}, u'Best Documentary, Short Subjects': {u'nominees': [[u'God Sleeps in Rwanda', u'0480247'], [u'The Death of Kevin Carter: Casualty of the Bang Bang Club', u'0439676'], [u'The Mushroom Club', u'0762119']], u'winner': [u'A Note of Triumph: The Golden Age of Norman Corwin', u'0491795']}, u'Best Sound': {u'nominees': [[u'The Chronicles of Narnia: The Lion, the Witch and the Wardrobe', u'0363771'], [u'Memoirs of a Geisha', u'0397535'], [u'Walk the Line', u'0358273'], [u'War of the Worlds', u'0407304']], u'winner': [u'King Kong', u'0360717']}, u'Best Director': {u'nominees': [[u'Capote', u'0379725'], [u'Crash', u'0375679'], [u'Good Night, and Good Luck.', u'0433383'], [u'Munich', u'0408306']], u'winner': [u'Brokeback Mountain', u'0388795']}, u'Best Actor in a Leading Role': {u'nominees': [[u'Brokeback Mountain', u'0388795'], [u'Good Night, and Good Luck.', u'0433383'], [u'Hustle & Flow', u'0410097'], [u'Walk the Line', u'0358273']], u'winner': [u'Capote', u'0379725']}, u'Best Picture': {u'nominees': [[u'Brokeback Mountain', u'0388795'], [u'Capote', u'0379725'], [u'Good Night, and Good Luck.', u'0433383'], [u'Munich', u'0408306']], u'winner': [u'Crash', u'0375679']}, u'Best Actress in a Leading Role': {u'nominees': [[u'Mrs Henderson Presents', u'0413015'], [u'North Country', u'0395972'], [u'Pride & Prejudice', u'0414387'], [u'Transamerica', u'0407265']], u'winner': [u'Walk the Line', u'0358273']}, u'Best Effects, Visual Effects': {u'nominees': [[u'The Chronicles of Narnia: The Lion, the Witch and the Wardrobe', u'0363771'], [u'War of the Worlds', u'0407304']], u'winner': [u'King Kong', u'0360717']}, u'Best Animated Feature Film of the Year': {u'nominees': [[u'Corpse Bride', u'0121164'], [u\"Howl's Moving Castle\", u'0347149']], u'winner': [u'The Curse of the Were-Rabbit', u'0312004']}}, u'2007': {u'Best Actor in a Supporting Role': {u'nominees': [[u'Blood Diamond', u'0450259'], [u'The Departed', u'0407887'], [u'Dreamgirls', u'0443489'], [u'Little Children', u'0404203']], u'winner': [u'Little Miss Sunshine', u'0449059']}, u'Best Documentary, Features': {u'nominees': [[u'P.O.V.', u'0795174'], [u'Deliver Us from Evil', u'0814075'], [u'Iraq in Fragments', u'0492466'], [u'Jesus Camp', u'0486358']], u'winner': [u'An Inconvenient Truth', u'0497116']}, u'Best Writing, Screenplay Written Directly for the Screen': {u'nominees': [[u'Babel', u'0449467'], [u\"Pan's Labyrinth\", u'0457430'], [u'Letters from Iwo Jima', u'0498380'], [u'The Queen', u'0436697']], u'winner': [u'Little Miss Sunshine', u'0449059']}, u'Best Foreign Language Film': {u'nominees': [[u'After the Wedding', u'0457655'], [u'Days of Glory', u'0444182'], [u\"Pan's Labyrinth\", u'0457430'], [u'Water', u'0240200']], u'winner': [u'The Lives of Others', u'0405094']}, u'Best Film Editing': {u'nominees': [[u'Babel', u'0449467'], [u'Blood Diamond', u'0450259'], [u'Children of Men', u'0206634'], [u'United 93', u'0475276']], u'winner': [u'The Departed', u'0407887']}, u'Best Art Direction': {u'nominees': [[u'Dreamgirls', u'0443489'], [u'The Good Shepherd', u'0343737'], [u\"Pirates of the Caribbean: Dead Man's Chest\", u'0383574'], [u'The Prestige', u'0482571']], u'winner': [u\"Pan's Labyrinth\", u'0457430']}, u'Best Short Film, Live Action': {u'nominees': [[u'Binta and the Great Idea', u'0442001'], [u'Helmer & Son', u'0955271'], [u'The Saviour', u'0467485'], [u'One Too Many', u'0460616']], u'winner': [u'West Bank Story', u'0438575']}, u'Best Music, Original Song': {u'nominees': [[u'Cars', u'0317219'], [u'Dreamgirls', u'0443489'], [u'Dreamgirls', u'0443489'], [u'Dreamgirls', u'0443489']], u'winner': [u'An Inconvenient Truth', u'0497116']}, u'Best Music, Original Score': {u'nominees': [[u'The Good German', u'0452624'], [u\"Pan's Labyrinth\", u'0457430'], [u'Notes on a Scandal', u'0465551'], [u'The Queen', u'0436697']], u'winner': [u'Babel', u'0449467']}, u'Best Short Film, Animated': {u'nominees': [[u'Lifted', u'0945571'], [u'The Little Matchgirl', u'0816562'], [u'Maestro', u'0951144'], [u'No Time for Nuts', u'0902999']], u'winner': [u'The Danish Poet', u'0933357']}, u'Best Makeup': {u'nominees': [[u'Apocalypto', u'0472043'], [u'Click', u'0389860']], u'winner': [u\"Pan's Labyrinth\", u'0457430']}, u'Best Costume Design': {u'nominees': [[u'The Devil Wears Prada', u'0458352'], [u'Dreamgirls', u'0443489'], [u'Curse of the Golden Flower', u'0473444'], [u'The Queen', u'0436697']], u'winner': [u'Marie Antoinette', u'0422720']}, u'Best Writing, Screenplay Based on Material from Another Medium': {u'nominees': [[u'Borat: Cultural Learnings of America for Make Benefit Glorious Nation of Kazakhstan', u'0443453'], [u'Children of Men', u'0206634'], [u'Little Children', u'0404203'], [u'Notes on a Scandal', u'0465551']], u'winner': [u'The Departed', u'0407887']}, u'Best Sound Editing': {u'nominees': [[u'Apocalypto', u'0472043'], [u'Blood Diamond', u'0450259'], [u'Flags of Our Fathers', u'0418689'], [u\"Pirates of the Caribbean: Dead Man's Chest\", u'0383574']], u'winner': [u'Letters from Iwo Jima', u'0498380']}, u'Best Actress in a Supporting Role': {u'nominees': [[u'Babel', u'0449467'], [u'Babel', u'0449467'], [u'Little Miss Sunshine', u'0449059'], [u'Notes on a Scandal', u'0465551']], u'winner': [u'Dreamgirls', u'0443489']}, u'Best Cinematography': {u'nominees': [[u'The Black Dahlia', u'0387877'], [u'Children of Men', u'0206634'], [u'The Illusionist', u'0443543'], [u'The Prestige', u'0482571']], u'winner': [u\"Pan's Labyrinth\", u'0457430']}, u'Best Documentary, Short Subjects': {u'nominees': [[u'Recycled Life', u'0859631'], [u'Rehearsing a Dream', u'0982919'], [u'Two Hands: The Leon Fleisher Story', u'0951151']], u'winner': [u'The Blood of Yingzhou District', u'0859595']}, u'Best Sound': {u'nominees': [[u'Apocalypto', u'0472043'], [u'Blood Diamond', u'0450259'], [u'Flags of Our Fathers', u'0418689'], [u\"Pirates of the Caribbean: Dead Man's Chest\", u'0383574']], u'winner': [u'Dreamgirls', u'0443489']}, u'Best Director': {u'nominees': [[u'Babel', u'0449467'], [u'Letters from Iwo Jima', u'0498380'], [u'The Queen', u'0436697'], [u'United 93', u'0475276']], u'winner': [u'The Departed', u'0407887']}, u'Best Actor in a Leading Role': {u'nominees': [[u'Blood Diamond', u'0450259'], [u'Half Nelson', u'0468489'], [u'The Pursuit of Happyness', u'0454921'], [u'Venus', u'0489327']], u'winner': [u'The Last King of Scotland', u'0455590']}, u'Best Picture': {u'nominees': [[u'Babel', u'0449467'], [u'Letters from Iwo Jima', u'0498380'], [u'Little Miss Sunshine', u'0449059'], [u'The Queen', u'0436697']], u'winner': [u'The Departed', u'0407887']}, u'Best Actress in a Leading Role': {u'nominees': [[u'The Devil Wears Prada', u'0458352'], [u'Little Children', u'0404203'], [u'Notes on a Scandal', u'0465551'], [u'Volver', u'0441909']], u'winner': [u'The Queen', u'0436697']}, u'Best Effects, Visual Effects': {u'nominees': [[u'Poseidon', u'0409182'], [u'Superman Returns', u'0348150']], u'winner': [u\"Pirates of the Caribbean: Dead Man's Chest\", u'0383574']}, u'Best Animated Feature Film of the Year': {u'nominees': [[u'Cars', u'0317219'], [u'Monster House', u'0385880']], u'winner': [u'Happy Feet', u'0366548']}}, u'2004': {u'Best Actor in a Supporting Role': {u'nominees': [[u'21 Grams', u'0315733'], [u'The Cooler', u'0318374'], [u'In America', u'0298845'], [u'The Last Samurai', u'0325710']], u'winner': [u'Mystic River', u'0327056']}, u'Best Documentary, Features': {u'nominees': [[u'Cuban Rafters', u'0321376'], [u'Capturing the Friedmans', u'0342172'], [u'My Architect', u'0373175'], [u'The Weather Underground', u'0343168']], u'winner': [u'The Fog of War: Eleven Lessons from the Life of Robert S. McNamara', u'0317910']}, u'Best Writing, Screenplay Written Directly for the Screen': {u'nominees': [[u'Dirty Pretty Things', u'0301199'], [u'Finding Nemo', u'0266543'], [u'In America', u'0298845'], [u'The Barbarian Invasions', u'0338135']], u'winner': [u'Lost in Translation', u'0335266']}, u'Best Foreign Language Film': {u'nominees': [[u'Evil', u'0338309'], [u'The Twilight Samurai', u'0351817'], [u'Twin Sisters', u'0322674'], [u'Zelary', u'0288330']], u'winner': [u'The Barbarian Invasions', u'0338135']}, u'Best Short Film, Live Action': {u'nominees': [[u'(A)Torzija', u'0350476'], [u'Most', u'0345672'], [u'The Red Jacket', u'0323844'], [u'Squash', u'0341588']], u'winner': [u'Two Soldiers', u'0304858']}, u'Best Music, Original Song': {u'nominees': [[u'Cold Mountain', u'0159365'], [u'Cold Mountain', u'0159365'], [u'A Mighty Wind', u'0310281'], [u'The Triplets of Belleville', u'0286244']], u'winner': [u'The Lord of the Rings: The Return of the King', u'0167260']}, u'Best Music, Original Score': {u'nominees': [[u'Big Fish', u'0319061'], [u'Cold Mountain', u'0159365'], [u'Finding Nemo', u'0266543'], [u'House of Sand and Fog', u'0315983']], u'winner': [u'The Lord of the Rings: The Return of the King', u'0167260']}, u'Best Art Direction-Set Decoration': {u'nominees': [[u'Girl with a Pearl Earring', u'0335119'], [u'The Last Samurai', u'0325710'], [u'Master and Commander: The Far Side of the World', u'0311113'], [u'Seabiscuit', u'0329575']], u'winner': [u'The Lord of the Rings: The Return of the King', u'0167260']}, u'Best Short Film, Animated': {u'nominees': [[u\"Boundin'\", u'0395479'], [u'Destino', u'0377770'], [u'Gone Nutty', u'0342965'], [u'Nibbles', u'0392542']], u'winner': [u'Harvie Krumpet', u'0382734']}, u'Best Makeup': {u'nominees': [[u'Master and Commander: The Far Side of the World', u'0311113'], [u'Pirates of the Caribbean: The Curse of the Black Pearl', u'0325980']], u'winner': [u'The Lord of the Rings: The Return of the King', u'0167260']}, u'Best Costume Design': {u'nominees': [[u'Girl with a Pearl Earring', u'0335119'], [u'The Last Samurai', u'0325710'], [u'Master and Commander: The Far Side of the World', u'0311113'], [u'Seabiscuit', u'0329575']], u'winner': [u'The Lord of the Rings: The Return of the King', u'0167260']}, u'Best Animated Feature': {u'nominees': [[u'Brother Bear', u'0328880'], [u'The Triplets of Belleville', u'0286244']], u'winner': [u'Finding Nemo', u'0266543']}, u'Best Writing, Screenplay Based on Material from Another Medium': {u'nominees': [[u'American Splendor', u'0305206'], [u'City of God', u'0317248'], [u'Mystic River', u'0327056'], [u'Seabiscuit', u'0329575']], u'winner': [u'The Lord of the Rings: The Return of the King', u'0167260']}, u'Best Sound Editing': {u'nominees': [[u'Finding Nemo', u'0266543'], [u'Pirates of the Caribbean: The Curse of the Black Pearl', u'0325980']], u'winner': [u'Master and Commander: The Far Side of the World', u'0311113']}, u'Best Actress in a Supporting Role': {u'nominees': [[u'House of Sand and Fog', u'0315983'], [u'Mystic River', u'0327056'], [u'Pieces of April', u'0311648'], [u'Thirteen', u'0328538']], u'winner': [u'Cold Mountain', u'0159365']}, u'Best Sound Mixing': {u'nominees': [[u'The Last Samurai', u'0325710'], [u'Master and Commander: The Far Side of the World', u'0311113'], [u'Pirates of the Caribbean: The Curse of the Black Pearl', u'0325980'], [u'Seabiscuit', u'0329575']], u'winner': [u'The Lord of the Rings: The Return of the King', u'0167260']}, u'Best Cinematography': {u'nominees': [[u'City of God', u'0317248'], [u'Cold Mountain', u'0159365'], [u'Girl with a Pearl Earring', u'0335119'], [u'Seabiscuit', u'0329575']], u'winner': [u'Master and Commander: The Far Side of the World', u'0311113']}, u'Best Documentary, Short Subjects': {u'nominees': [[u'Asylum', u'0363445'], [u'Ferry Tales', u'0363607']], u'winner': [u'Chernobyl Heart', u'0396959']}, u'Best Film Editing': {u'nominees': [[u'City of God', u'0317248'], [u'Cold Mountain', u'0159365'], [u'Master and Commander: The Far Side of the World', u'0311113'], [u'Seabiscuit', u'0329575']], u'winner': [u'The Lord of the Rings: The Return of the King', u'0167260']}, u'Best Director': {u'nominees': [[u'City of God', u'0317248'], [u'Lost in Translation', u'0335266'], [u'Master and Commander: The Far Side of the World', u'0311113'], [u'Mystic River', u'0327056']], u'winner': [u'The Lord of the Rings: The Return of the King', u'0167260']}, u'Best Actor in a Leading Role': {u'nominees': [[u'Cold Mountain', u'0159365'], [u'House of Sand and Fog', u'0315983'], [u'Lost in Translation', u'0335266'], [u'Pirates of the Caribbean: The Curse of the Black Pearl', u'0325980']], u'winner': [u'Mystic River', u'0327056']}, u'Best Picture': {u'nominees': [[u'Lost in Translation', u'0335266'], [u'Master and Commander: The Far Side of the World', u'0311113'], [u'Mystic River', u'0327056'], [u'Seabiscuit', u'0329575']], u'winner': [u'The Lord of the Rings: The Return of the King', u'0167260']}, u'Best Actress in a Leading Role': {u'nominees': [[u'Whale Rider', u'0298228'], [u\"Something's Gotta Give\", u'0337741'], [u'In America', u'0298845'], [u'21 Grams', u'0315733']], u'winner': [u'Monster', u'0340855']}, u'Best Visual Effects': {u'nominees': [[u'Master and Commander: The Far Side of the World', u'0311113'], [u'Pirates of the Caribbean: The Curse of the Black Pearl', u'0325980']], u'winner': [u'The Lord of the Rings: The Return of the King', u'0167260']}}, u'2005': {u'Best Actor in a Supporting Role': {u'nominees': [[u'The Aviator', u'0338751'], [u'Closer', u'0376541'], [u'Collateral', u'0369339'], [u'Sideways', u'0375063']], u'winner': [u'Million Dollar Baby', u'0405159']}, u'Best Documentary, Features': {u'nominees': [[u'The Story of the Weeping Camel', u'0373861'], [u'Super Size Me', u'0390521'], [u'Tupac: Resurrection', u'0343121'], [u'Twist of Faith', u'0436857']], u'winner': [u\"Born Into Brothels: Calcutta's Red Light Kids\", u'0388789']}, u'Best Writing, Screenplay Written Directly for the Screen': {u'nominees': [[u'The Aviator', u'0338751'], [u'Hotel Rwanda', u'0395169'], [u'The Incredibles', u'0317705'], [u'Vera Drake', u'0383694']], u'winner': [u'Eternal Sunshine of the Spotless Mind', u'0338013']}, u'Best Foreign Language Film': {u'nominees': [[u'The Chorus', u'0372824'], [u'As It Is in Heaven', u'0382330'], [u'Downfall', u'0363163'], [u'Yesterday', u'0419279']], u'winner': [u'The Sea Inside', u'0369702']}, u'Best Film Editing': {u'nominees': [[u'Collateral', u'0369339'], [u'Finding Neverland', u'0308644'], [u'Million Dollar Baby', u'0405159'], [u'Ray', u'0350258']], u'winner': [u'The Aviator', u'0338751']}, u'Best Art Direction': {u'nominees': [[u'Finding Neverland', u'0308644'], [u\"Lemony Snicket's A Series of Unfortunate Events\", u'0339291'], [u'The Phantom of the Opera', u'0293508'], [u'A Very Long Engagement', u'0344510']], u'winner': [u'The Aviator', u'0338751']}, u'Best Short Film, Live Action': {u'nominees': [[u'7:35 in the Morning', u'0406501'], [u'Everything in This Country Must', u'0443499'], [u'Little Terrorist', u'0425200'], [u'Two Cars, One Night', u'0390579']], u'winner': [u'Wasp', u'0388534']}, u'Best Music, Original Song': {u'nominees': [[u'The Chorus', u'0372824'], [u'The Phantom of the Opera', u'0293508'], [u'The Polar Express', u'0338348'], [u'Shrek 2', u'0298148']], u'winner': [u'The Motorcycle Diaries', u'0318462']}, u'Best Music, Original Score': {u'nominees': [[u'Harry Potter and the Prisoner of Azkaban', u'0304141'], [u\"Lemony Snicket's A Series of Unfortunate Events\", u'0339291'], [u'The Passion of the Christ', u'0335345'], [u'The Village', u'0368447']], u'winner': [u'Finding Neverland', u'0308644']}, u'Best Short Film, Animated': {u'nominees': [[u'Birthday Boy', u'0435933'], [u'Gopher Broke', u'0438064'], [u'Guard Dog', u'0430159'], [u'Lorenzo', u'0412936']], u'winner': [u'Ryan', u'0414469']}, u'Best Makeup': {u'nominees': [[u'The Sea Inside', u'0369702'], [u'The Passion of the Christ', u'0335345']], u'winner': [u\"Lemony Snicket's A Series of Unfortunate Events\", u'0339291']}, u'Best Costume Design': {u'nominees': [[u'Finding Neverland', u'0308644'], [u\"Lemony Snicket's A Series of Unfortunate Events\", u'0339291'], [u'Ray', u'0350258'], [u'Troy', u'0332452']], u'winner': [u'The Aviator', u'0338751']}, u'Best Writing, Screenplay Based on Material from Another Medium': {u'nominees': [[u'Before Sunset', u'0381681'], [u'The Motorcycle Diaries', u'0318462'], [u'Finding Neverland', u'0308644'], [u'Million Dollar Baby', u'0405159']], u'winner': [u'Sideways', u'0375063']}, u'Best Sound Editing': {u'nominees': [[u'The Polar Express', u'0338348'], [u'Spider-Man 2', u'0316654']], u'winner': [u'The Incredibles', u'0317705']}, u'Best Actress in a Supporting Role': {u'nominees': [[u'Closer', u'0376541'], [u'Hotel Rwanda', u'0395169'], [u'Kinsey', u'0362269'], [u'Sideways', u'0375063']], u'winner': [u'The Aviator', u'0338751']}, u'Best Cinematography': {u'nominees': [[u'The Passion of the Christ', u'0335345'], [u'The Phantom of the Opera', u'0293508'], [u'House of Flying Daggers', u'0385004'], [u'A Very Long Engagement', u'0344510']], u'winner': [u'The Aviator', u'0338751']}, u'Best Documentary, Short Subjects': {u'nominees': [[u'Autism Is a World', u'0443434'], [u'The Children of Leningradsky', u'0442947'], [u'Hardwood', u'0443531'], [u\"Sister Rose's Passion\", u'0420195']], u'winner': [u\"Mighty Times: The Children's March\", u'0443587']}, u'Best Sound': {u'nominees': [[u'The Aviator', u'0338751'], [u'The Incredibles', u'0317705'], [u'The Polar Express', u'0338348'], [u'Spider-Man 2', u'0316654']], u'winner': [u'Ray', u'0350258']}, u'Best Director': {u'nominees': [[u'The Aviator', u'0338751'], [u'Ray', u'0350258'], [u'Sideways', u'0375063'], [u'Vera Drake', u'0383694']], u'winner': [u'Million Dollar Baby', u'0405159']}, u'Best Actor in a Leading Role': {u'nominees': [[u'The Aviator', u'0338751'], [u'Finding Neverland', u'0308644'], [u'Hotel Rwanda', u'0395169'], [u'Million Dollar Baby', u'0405159']], u'winner': [u'Ray', u'0350258']}, u'Best Picture': {u'nominees': [[u'The Aviator', u'0338751'], [u'Finding Neverland', u'0308644'], [u'Ray', u'0350258'], [u'Sideways', u'0375063']], u'winner': [u'Million Dollar Baby', u'0405159']}, u'Best Actress in a Leading Role': {u'nominees': [[u'Being Julia', u'0340012'], [u'Eternal Sunshine of the Spotless Mind', u'0338013'], [u'Maria Full of Grace', u'0390221'], [u'Vera Drake', u'0383694']], u'winner': [u'Million Dollar Baby', u'0405159']}, u'Best Effects, Visual Effects': {u'nominees': [[u'Harry Potter and the Prisoner of Azkaban', u'0304141'], [u'I, Robot', u'0343818']], u'winner': [u'Spider-Man 2', u'0316654']}, u'Best Animated Feature Film of the Year': {u'nominees': [[u'Shark Tale', u'0307453'], [u'Shrek 2', u'0298148']], u'winner': [u'The Incredibles', u'0317705']}}, u'2008': {u'Best Actor in a Supporting Role': {u'nominees': [[u'The Assassination of Jesse James by the Coward Robert Ford', u'0443680'], [u\"Charlie Wilson's War\", u'0472062'], [u'Into the Wild', u'0758758'], [u'Michael Clayton', u'0465538']], u'winner': [u'No Country for Old Men', u'0477348']}, u'Best Documentary, Features': {u'nominees': [[u'No End in Sight', u'0912593'], [u'Operation Homecoming: Writing the Wartime Experience', u'0970184'], [u'Sicko', u'0386032'], [u'War Dance', u'0912599']], u'winner': [u'Taxi to the Dark Side', u'0854678']}, u'Best Writing, Screenplay Written Directly for the Screen': {u'nominees': [[u'Lars and the Real Girl', u'0805564'], [u'Michael Clayton', u'0465538'], [u'Ratatouille', u'0382932'], [u'The Savages', u'0775529']], u'winner': [u'Juno', u'0467406']}, u'Best Foreign Language Film': {u'nominees': [[u'Beaufort', u'0758732'], [u'Mongol: The Rise of Genghis Khan', u'0416044'], [u'Katyn', u'0879843'], [u'12', u'0488478']], u'winner': [u'The Counterfeiters', u'0813547']}, u'Best Film Editing': {u'nominees': [[u'Into the Wild', u'0758758'], [u'No Country for Old Men', u'0477348'], [u'The Diving Bell and the Butterfly', u'0401383'], [u'There Will Be Blood', u'0469494']], u'winner': [u'The Bourne Ultimatum', u'0440963']}, u'Best Art Direction': {u'nominees': [[u'American Gangster', u'0765429'], [u'Atonement', u'0783233'], [u'The Golden Compass', u'0385752'], [u'There Will Be Blood', u'0469494']], u'winner': [u'Sweeney Todd: The Demon Barber of Fleet Street', u'0408236']}, u'Best Short Film, Live Action': {u'nominees': [[u'At Night', u'1117985'], [u'The Substitute', u'0922627'], [u'Tanghi argentini', u'0812387'], [u'The Tonto Woman', u'1176466']], u'winner': [u'Le Mozart des pickpockets', u'0820971']}, u'Best Music, Original Song': {u'nominees': [[u'August Rush', u'0426931'], [u'Enchanted', u'0461770'], [u'Enchanted', u'0461770'], [u'Enchanted', u'0461770']], u'winner': [u'Once', u'0907657']}, u'Best Music, Original Score': {u'nominees': [[u'3:10 to Yuma', u'0381849'], [u'The Kite Runner', u'0419887'], [u'Michael Clayton', u'0465538'], [u'Ratatouille', u'0382932']], u'winner': [u'Atonement', u'0783233']}, u'Best Short Film, Animated': {u'nominees': [[u'I Met the Walrus', u'1156511'], [u'Madame Tutli-Putli', u'1029440'], [u'My Love', u'0874952'], [u'Even Pigeons Go to Heaven', u'1066329']], u'winner': [u'Peter & the Wolf', u'0863136']}, u'Best Makeup': {u'nominees': [[u'Norbit', u'0477051'], [u\"Pirates of the Caribbean: At World's End\", u'0449088']], u'winner': [u'La Vie en Rose', u'0450188']}, u'Best Costume Design': {u'nominees': [[u'Across the Universe', u'0445922'], [u'Atonement', u'0783233'], [u'La Vie en Rose', u'0450188'], [u'Sweeney Todd: The Demon Barber of Fleet Street', u'0408236']], u'winner': [u'Elizabeth: The Golden Age', u'0414055']}, u'Best Writing, Screenplay Based on Material from Another Medium': {u'nominees': [[u'Atonement', u'0783233'], [u'Away from Her', u'0491747'], [u'The Diving Bell and the Butterfly', u'0401383'], [u'There Will Be Blood', u'0469494']], u'winner': [u'No Country for Old Men', u'0477348']}, u'Best Sound Editing': {u'nominees': [[u'No Country for Old Men', u'0477348'], [u'Ratatouille', u'0382932'], [u'There Will Be Blood', u'0469494'], [u'Transformers', u'0418279']], u'winner': [u'The Bourne Ultimatum', u'0440963']}, u'Best Actress in a Supporting Role': {u'nominees': [[u'American Gangster', u'0765429'], [u'Atonement', u'0783233'], [u'Gone Baby Gone', u'0452623'], [u\"I'm Not There.\", u'0368794']], u'winner': [u'Michael Clayton', u'0465538']}, u'Best Cinematography': {u'nominees': [[u'The Assassination of Jesse James by the Coward Robert Ford', u'0443680'], [u'Atonement', u'0783233'], [u'No Country for Old Men', u'0477348'], [u'The Diving Bell and the Butterfly', u'0401383']], u'winner': [u'There Will Be Blood', u'0469494']}, u'Best Documentary, Short Subjects': {u'nominees': [[u'La corona', u'1157626'], [u'Salim Baba', u'1157707'], [u\"Sari's Mother\", u'0841160']], u'winner': [u'Freeheld', u'0914369']}, u'Best Sound': {u'nominees': [[u'3:10 to Yuma', u'0381849'], [u'No Country for Old Men', u'0477348'], [u'Ratatouille', u'0382932'], [u'Transformers', u'0418279']], u'winner': [u'The Bourne Ultimatum', u'0440963']}, u'Best Director': {u'nominees': [[u'Juno', u'0467406'], [u'Michael Clayton', u'0465538'], [u'The Diving Bell and the Butterfly', u'0401383'], [u'There Will Be Blood', u'0469494']], u'winner': [u'No Country for Old Men', u'0477348']}, u'Best Actor in a Leading Role': {u'nominees': [[u'Eastern Promises', u'0765443'], [u'In the Valley of Elah', u'0478134'], [u'Michael Clayton', u'0465538'], [u'Sweeney Todd: The Demon Barber of Fleet Street', u'0408236']], u'winner': [u'There Will Be Blood', u'0469494']}, u'Best Picture': {u'nominees': [[u'Atonement', u'0783233'], [u'Juno', u'0467406'], [u'Michael Clayton', u'0465538'], [u'There Will Be Blood', u'0469494']], u'winner': [u'No Country for Old Men', u'0477348']}, u'Best Actress in a Leading Role': {u'nominees': [[u'Away from Her', u'0491747'], [u'Elizabeth: The Golden Age', u'0414055'], [u'Juno', u'0467406'], [u'The Savages', u'0775529']], u'winner': [u'La Vie en Rose', u'0450188']}, u'Best Effects, Visual Effects': {u'nominees': [[u\"Pirates of the Caribbean: At World's End\", u'0449088'], [u'Transformers', u'0418279']], u'winner': [u'The Golden Compass', u'0385752']}, u'Best Animated Feature Film of the Year': {u'nominees': [[u'Persepolis', u'0808417'], [u\"Surf's Up\", u'0423294']], u'winner': [u'Ratatouille', u'0382932']}}, u'2009': {u'Best Actor in a Supporting Role': {u'nominees': [[u'Doubt', u'0918927'], [u'Milk', u'1013753'], [u'Revolutionary Road', u'0959337'], [u'Tropic Thunder', u'0942385']], u'winner': [u'The Dark Knight', u'0468569']}, u'Best Documentary, Features': {u'nominees': [[u'The Betrayal - Nerakhoon', u'1157685'], [u'Encounters at the End of the World', u'1093824'], [u'The Garden', u'1252486'], [u'Trouble the Water', u'1149405']], u'winner': [u'Man on Wire', u'1155592']}, u'Best Writing, Screenplay Written Directly for the Screen': {u'nominees': [[u'Frozen River', u'0978759'], [u'Happy-Go-Lucky', u'1045670'], [u'In Bruges', u'0780536'], [u'WALLeE', u'0910970']], u'winner': [u'Milk', u'1013753']}, u'Best Foreign Language Film': {u'nominees': [[u'The Class', u'1068646'], [u'The Baader Meinhof Complex', u'0765432'], [u'Waltz with Bashir', u'1185616'], [u'Revanche', u'1173745']], u'winner': [u'Departures', u'1069238']}, u'Best Film Editing': {u'nominees': [[u'The Curious Case of Benjamin Button', u'0421715'], [u'The Dark Knight', u'0468569'], [u'Frost/Nixon', u'0870111'], [u'Milk', u'1013753']], u'winner': [u'Slumdog Millionaire', u'1010048']}, u'Best Art Direction': {u'nominees': [[u'Changeling', u'0824747'], [u'The Dark Knight', u'0468569'], [u'The Duchess', u'0864761'], [u'Revolutionary Road', u'0959337']], u'winner': [u'The Curious Case of Benjamin Button', u'0421715']}, u'Best Documentary, Short Subject': {u'nominees': [[u'The Conscience of Nhem En', u'1360057'], [u'The Final Inch', u'1360061'], [u'The Witness from the Balcony of Room 306', u'1360068']], u'winner': [u'Smile Pinki', u'1360065']}, u'Best Short Film, Live Action': {u'nominees': [[u'On the Line', u'1219675'], [u'The Pig', u'1333640'], [u'Manon on the Asphalt', u'1186020'], [u'New Boy', u'1095191']], u'winner': [u'Toyland', u'1280548']}, u'Best Music, Original Score': {u'nominees': [[u'The Curious Case of Benjamin Button', u'0421715'], [u'Defiance', u'1034303'], [u'Milk', u'1013753'], [u'WALLeE', u'0910970']], u'winner': [u'Slumdog Millionaire', u'1010048']}, u'Best Short Film, Animated': {u'nominees': [[u'Oktapodi', u'1323941'], [u'Presto', u'1245104'], [u'This Way Up', u'1279499'], [u'Lavatory Lovestory', u'0983242']], u'winner': [u'The House of Small Cubes', u'1361566']}, u'Best Makeup': {u'nominees': [[u'The Dark Knight', u'0468569'], [u'Hellboy II: The Golden Army', u'0411477']], u'winner': [u'The Curious Case of Benjamin Button', u'0421715']}, u'Best Costume Design': {u'nominees': [[u'Australia', u'0455824'], [u'The Curious Case of Benjamin Button', u'0421715'], [u'Milk', u'1013753'], [u'Revolutionary Road', u'0959337']], u'winner': [u'The Duchess', u'0864761']}, u'Best Writing, Screenplay Based on Material from Another Medium': {u'nominees': [[u'The Curious Case of Benjamin Button', u'0421715'], [u'Doubt', u'0918927'], [u'Frost/Nixon', u'0870111'], [u'The Reader', u'0976051']], u'winner': [u'Slumdog Millionaire', u'1010048']}, u'Best Sound Editing': {u'nominees': [[u'Iron Man', u'0371746'], [u'Slumdog Millionaire', u'1010048'], [u'WALLeE', u'0910970'], [u'Wanted', u'0493464']], u'winner': [u'The Dark Knight', u'0468569']}, u'Best Actress in a Supporting Role': {u'nominees': [[u'The Curious Case of Benjamin Button', u'0421715'], [u'Doubt', u'0918927'], [u'Doubt', u'0918927'], [u'The Wrestler', u'1125849']], u'winner': [u'Vicky Cristina Barcelona', u'0497465']}, u'Best Cinematography': {u'nominees': [[u'Changeling', u'0824747'], [u'The Curious Case of Benjamin Button', u'0421715'], [u'The Dark Knight', u'0468569'], [u'The Reader', u'0976051']], u'winner': [u'Slumdog Millionaire', u'1010048']}, u'Best Music, Original Song': {u'nominees': [[u'Slumdog Millionaire', u'1010048'], [u'WALLeE', u'0910970']], u'winner': [u'Slumdog Millionaire', u'1010048']}, u'Best Sound': {u'nominees': [[u'The Curious Case of Benjamin Button', u'0421715'], [u'The Dark Knight', u'0468569'], [u'WALLeE', u'0910970'], [u'Wanted', u'0493464']], u'winner': [u'Slumdog Millionaire', u'1010048']}, u'Best Director': {u'nominees': [[u'The Curious Case of Benjamin Button', u'0421715'], [u'Frost/Nixon', u'0870111'], [u'Milk', u'1013753'], [u'The Reader', u'0976051']], u'winner': [u'Slumdog Millionaire', u'1010048']}, u'Best Actor in a Leading Role': {u'nominees': [[u'The Curious Case of Benjamin Button', u'0421715'], [u'Frost/Nixon', u'0870111'], [u'The Visitor', u'0857191'], [u'The Wrestler', u'1125849']], u'winner': [u'Milk', u'1013753']}, u'Best Picture': {u'nominees': [[u'The Curious Case of Benjamin Button', u'0421715'], [u'Frost/Nixon', u'0870111'], [u'Milk', u'1013753'], [u'The Reader', u'0976051']], u'winner': [u'Slumdog Millionaire', u'1010048']}, u'Best Actress in a Leading Role': {u'nominees': [[u'Changeling', u'0824747'], [u'Doubt', u'0918927'], [u'Frozen River', u'0978759'], [u'Rachel Getting Married', u'1084950']], u'winner': [u'The Reader', u'0976051']}, u'Best Effects, Visual Effects': {u'nominees': [[u'The Dark Knight', u'0468569'], [u'Iron Man', u'0371746']], u'winner': [u'The Curious Case of Benjamin Button', u'0421715']}, u'Best Animated Feature Film of the Year': {u'nominees': [[u'Bolt', u'0397892'], [u'Kung Fu Panda', u'0441773']], u'winner': [u'WALLeE', u'0910970']}}, u'2011': {u'Best Actor in a Supporting Role': {u'nominees': [[u\"Winter's Bone\", u'1399683'], [u'The Town', u'0840361'], [u'The Kids Are All Right', u'0842926'], [u\"The King's Speech\", u'1504320']], u'winner': [u'The Fighter', u'0964517']}, u'Best Documentary, Features': {u'nominees': [[u'Exit Through the Gift Shop', u'1587707'], [u'GasLand', u'1558250'], [u'Restrepo', u'1559549'], [u'Waste Land', u'1268204']], u'winner': [u'Inside Job', u'1645089']}, u'Best Writing, Screenplay Written Directly for the Screen': {u'nominees': [[u'Another Year', u'1431181'], [u'Inception', u'1375666'], [u'The Kids Are All Right', u'0842926'], [u'The Fighter', u'0964517']], u'winner': [u\"The King's Speech\", u'1504320']}, u'Best Foreign Language Film': {u'nominees': [[u'Biutiful', u'1164999'], [u'Dogtooth', u'1379182'], [u'Incendies', u'1255953'], [u'Outside the Law', u'1229381']], u'winner': [u'In a Better World', u'1340107']}, u'Best Film Editing': {u'nominees': [[u'127 Hours', u'1542344'], [u'Black Swan', u'0947798'], [u'The Fighter', u'0964517'], [u\"The King's Speech\", u'1504320']], u'winner': [u'The Social Network', u'1285016']}, u'Best Art Direction': {u'nominees': [[u'Harry Potter and the Deathly Hallows: Part 1', u'0926084'], [u'Inception', u'1375666'], [u\"The King's Speech\", u'1504320'], [u'True Grit', u'1403865']], u'winner': [u'Alice in Wonderland', u'1014759']}, u'Best Short Film, Live Action': {u'nominees': [[u'The Confession', u'1687903'], [u'The Crush', u'1509756'], [u'Na Wewe', u'1764508'], [u'Wish 143', u'1518330']], u'winner': [u'God of Love', u'1631323']}, u'Best Music, Original Song': {u'nominees': [[u'127 Hours', u'1542344'], [u'Country Strong', u'1555064'], [u'Tangled', u'0398286']], u'winner': [u'Toy Story 3', u'0435761']}, u'Best Music, Original Score': {u'nominees': [[u'127 Hours', u'1542344'], [u'How to Train Your Dragon', u'0892769'], [u'Inception', u'1375666'], [u\"The King's Speech\", u'1504320']], u'winner': [u'The Social Network', u'1285016']}, u'Best Short Film, Animated': {u'nominees': [[u'Day & Night', u'1620446'], [u'The Gruffalo', u'1461418'], [u\"Let's Pollute\", u'1470652'], [u'Madagascar, a Journey Diary', u'1590114']], u'winner': [u'The Lost Thing', u'1669698']}, u'Best Makeup': {u'nominees': [[u\"Barney's Version\", u'1423894'], [u'The Way Back', u'1023114']], u'winner': [u'The Wolfman', u'0780653']}, u'Best Costume Design': {u'nominees': [[u'I Am Love', u'1226236'], [u\"The King's Speech\", u'1504320'], [u'The Tempest', u'1274300'], [u'True Grit', u'1403865']], u'winner': [u'Alice in Wonderland', u'1014759']}, u'Best Writing, Screenplay Based on Material from Another Medium': {u'nominees': [[u'127 Hours', u'1542344'], [u'True Grit', u'1403865'], [u\"Winter's Bone\", u'1399683'], [u'Toy Story 3', u'0435761']], u'winner': [u'The Social Network', u'1285016']}, u'Best Sound Editing': {u'nominees': [[u'Toy Story 3', u'0435761'], [u'TRON: Legacy', u'1104001'], [u'True Grit', u'1403865'], [u'Unstoppable', u'0477080']], u'winner': [u'Inception', u'1375666']}, u'Best Actress in a Supporting Role': {u'nominees': [[u'The Fighter', u'0964517'], [u\"The King's Speech\", u'1504320'], [u'True Grit', u'1403865'], [u'Animal Kingdom', u'1313092']], u'winner': [u'The Fighter', u'0964517']}, u'Best Cinematography': {u'nominees': [[u'Black Swan', u'0947798'], [u\"The King's Speech\", u'1504320'], [u'The Social Network', u'1285016'], [u'True Grit', u'1403865']], u'winner': [u'Inception', u'1375666']}, u'Best Documentary, Short Subjects': {u'nominees': [[u'Poster Girl', u'1505384'], [u'Killing in the Name', u'1684897'], [u'Sun Come Up', u'1621975'], [u'The Warriors of Qiugang', u'1744887']], u'winner': [u'Strangers No More', u'1754549']}, u'Best Sound': {u'nominees': [[u\"The King's Speech\", u'1504320'], [u'Salt', u'0944835'], [u'The Social Network', u'1285016'], [u'True Grit', u'1403865']], u'winner': [u'Inception', u'1375666']}, u'Best Director': {u'nominees': [[u'Black Swan', u'0947798'], [u'True Grit', u'1403865'], [u'The Social Network', u'1285016'], [u'The Fighter', u'0964517']], u'winner': [u\"The King's Speech\", u'1504320']}, u'Best Actor in a Leading Role': {u'nominees': [[u'Biutiful', u'1164999'], [u'True Grit', u'1403865'], [u'The Social Network', u'1285016'], [u'127 Hours', u'1542344']], u'winner': [u\"The King's Speech\", u'1504320']}, u'Best Picture': {u'nominees': [[u'127 Hours', u'1542344'], [u'Black Swan', u'0947798'], [u'The Fighter', u'0964517'], [u'Inception', u'1375666'], [u'The Kids Are All Right', u'0842926'], [u'The Social Network', u'1285016'], [u'Toy Story 3', u'0435761'], [u'True Grit', u'1403865'], [u\"Winter's Bone\", u'1399683']], u'winner': [u\"The King's Speech\", u'1504320']}, u'Best Actress in a Leading Role': {u'nominees': [[u'The Kids Are All Right', u'0842926'], [u'Rabbit Hole', u'0935075'], [u\"Winter's Bone\", u'1399683'], [u'Blue Valentine', u'1120985']], u'winner': [u'Black Swan', u'0947798']}, u'Best Effects, Visual Effects': {u'nominees': [[u'Alice in Wonderland', u'1014759'], [u'Harry Potter and the Deathly Hallows: Part 1', u'0926084'], [u'Hereafter', u'1212419'], [u'Iron Man 2', u'1228705']], u'winner': [u'Inception', u'1375666']}, u'Best Animated Feature Film of the Year': {u'nominees': [[u'How to Train Your Dragon', u'0892769'], [u'The Illusionist', u'0775489']], u'winner': [u'Toy Story 3', u'0435761']}}, u'2010': {u'Best Actor in a Supporting Role': {u'nominees': [[u'Invictus', u'1057500'], [u'The Last Station', u'0824758'], [u'The Lovely Bones', u'0380510'], [u'The Messenger', u'0790712']], u'winner': [u'Inglourious Basterds', u'0361748']}, u'Best Documentary, Features': {u'nominees': [[u'Burma VJ: Reporter i et lukket land', u'1333634'], [u'Food, Inc.', u'1286537'], [u'The Most Dangerous Man in America: Daniel Ellsberg and the Pentagon Papers', u'1319726'], [u'Which Way Home', u'0489342']], u'winner': [u'The Cove', u'1313104']}, u'Best Writing, Screenplay Written Directly for the Screen': {u'nominees': [[u'A Serious Man', u'1019452'], [u'Inglourious Basterds', u'0361748'], [u'The Messenger', u'0790712'], [u'Up', u'1049413']], u'winner': [u'The Hurt Locker', u'0887912']}, u'Best Foreign Language Film': {u'nominees': [[u'The Milk of Sorrow', u'1206488'], [u'The White Ribbon', u'1149362'], [u'Ajami', u'1077262'], [u'A Prophet', u'1235166']], u'winner': [u'The Secret in Their Eyes', u'1305806']}, u'Best Film Editing': {u'nominees': [[u'Avatar', u'0499549'], [u'District 9', u'1136608'], [u'Inglourious Basterds', u'0361748'], [u'Precious', u'0929632']], u'winner': [u'The Hurt Locker', u'0887912']}, u'Best Art Direction': {u'nominees': [[u'Nine', u'0875034'], [u'Sherlock Holmes', u'0988045'], [u'The Imaginarium of Doctor Parnassus', u'1054606'], [u'The Young Victoria', u'0962736']], u'winner': [u'Avatar', u'0499549']}, u'Best Short Film, Live Action': {u'nominees': [[u'Instead of Abracadabra', u'1351684'], [u'Kavi', u'1399573'], [u'Miracle Fish', u'1337354'], [u'The Door', u'1381557']], u'winner': [u'The New Tenants', u'1520428']}, u'Best Music, Original Song': {u'nominees': [[u'Paris 36', u'0948535'], [u'Nine', u'0875034'], [u'The Princess and the Frog', u'0780521'], [u'The Princess and the Frog', u'0780521']], u'winner': [u'Crazy Heart', u'1263670']}, u'Best Music, Original Score': {u'nominees': [[u'Avatar', u'0499549'], [u'Fantastic Mr. Fox', u'0432283'], [u'Sherlock Holmes', u'0988045'], [u'The Hurt Locker', u'0887912']], u'winner': [u'Up', u'1049413']}, u'Best Short Film, Animated': {u'nominees': [[u'French Roast', u'1323520'], [u\"Granny O'Grimm's Sleeping Beauty\", u'1382454'], [u'The Lady and the Reaper (La dama y la muerte)', u'1523317'], [u'A Matter of Loaf and Death', u'1118511']], u'winner': [u'Logorama', u'1563725']}, u'Best Makeup': {u'nominees': [[u'Il divo: La spettacolare vita di Giulio Andreotti', u'1023490'], [u'The Young Victoria', u'0962736']], u'winner': [u'Star Trek', u'0796366']}, u'Best Costume Design': {u'nominees': [[u'Bright Star', u'0810784'], [u'Coco Before Chanel', u'1035736'], [u'Nine', u'0875034'], [u'The Imaginarium of Doctor Parnassus', u'1054606']], u'winner': [u'The Young Victoria', u'0962736']}, u'Best Writing, Screenplay Based on Material from Another Medium': {u'nominees': [[u'An Education', u'1174732'], [u'District 9', u'1136608'], [u'In the Loop', u'1226774'], [u'Up in the Air', u'1193138']], u'winner': [u'Precious', u'0929632']}, u'Best Sound Editing': {u'nominees': [[u'Avatar', u'0499549'], [u'Inglourious Basterds', u'0361748'], [u'Star Trek', u'0796366'], [u'Up', u'1049413']], u'winner': [u'The Hurt Locker', u'0887912']}, u'Best Actress in a Supporting Role': {u'nominees': [[u'Crazy Heart', u'1263670'], [u'Nine', u'0875034'], [u'Up in the Air', u'1193138'], [u'Up in the Air', u'1193138']], u'winner': [u'Precious', u'0929632']}, u'Best Cinematography': {u'nominees': [[u'The White Ribbon', u'1149362'], [u'Harry Potter and the Half-Blood Prince', u'0417741'], [u'Inglourious Basterds', u'0361748'], [u'The Hurt Locker', u'0887912']], u'winner': [u'Avatar', u'0499549']}, u'Best Documentary, Short Subjects': {u'nominees': [[u\"China's Unnatural Disaster: The Tears of Sichuan Province\", u'1430809'], [u'Rabbit e la Berlin', u'1410281'], [u'The Last Campaign of Governor Booth Gardner', u'1534853'], [u'The Last Truck: Closing of a GM Plant', u'1487120']], u'winner': [u'Music by Prudence', u'1543539']}, u'Best Sound': {u'nominees': [[u'Avatar', u'0499549'], [u'Inglourious Basterds', u'0361748'], [u'Star Trek', u'0796366'], [u'Transformers: Revenge of the Fallen', u'1055369']], u'winner': [u'The Hurt Locker', u'0887912']}, u'Best Director': {u'nominees': [[u'Avatar', u'0499549'], [u'Inglourious Basterds', u'0361748'], [u'Precious', u'0929632'], [u'Up in the Air', u'1193138']], u'winner': [u'The Hurt Locker', u'0887912']}, u'Best Actor in a Leading Role': {u'nominees': [[u'A Single Man', u'1315981'], [u'Invictus', u'1057500'], [u'The Hurt Locker', u'0887912'], [u'Up in the Air', u'1193138']], u'winner': [u'Crazy Heart', u'1263670']}, u'Best Picture': {u'nominees': [[u'A Serious Man', u'1019452'], [u'An Education', u'1174732'], [u'Avatar', u'0499549'], [u'District 9', u'1136608'], [u'Inglourious Basterds', u'0361748'], [u'Precious', u'0929632'], [u'The Blind Side', u'0878804'], [u'Up', u'1049413'], [u'Up in the Air', u'1193138']], u'winner': [u'The Hurt Locker', u'0887912']}, u'Best Actress in a Leading Role': {u'nominees': [[u'An Education', u'1174732'], [u'Julie & Julia', u'1135503'], [u'Precious', u'0929632'], [u'The Last Station', u'0824758']], u'winner': [u'The Blind Side', u'0878804']}, u'Best Effects, Visual Effects': {u'nominees': [[u'District 9', u'1136608'], [u'Star Trek', u'0796366']], u'winner': [u'Avatar', u'0499549']}, u'Best Animated Feature Film of the Year': {u'nominees': [[u'Coraline', u'0327597'], [u'Fantastic Mr. Fox', u'0432283'], [u'The Princess and the Frog', u'0780521'], [u'The Secret of Kells', u'0485601']], u'winner': [u'Up', u'1049413']}}, u'2013': {u'Best Actor in a Supporting Role': {u'nominees': [[u'Argo', u'1024648'], [u'Silver Linings Playbook', u'1045658'], [u'The Master', u'1560747'], [u'Lincoln', u'0443272']], u'winner': [u'Django Unchained', u'1853728']}, u'Best Documentary, Features': {u'nominees': [[u'5 Broken Cameras', u'2125423'], [u'The Gatekeepers', u'2309788'], [u'How to Survive a Plague', u'2124803'], [u'The Invisible War', u'2120152']], u'winner': [u'Searching for Sugar Man', u'2125608']}, u'Best Writing, Screenplay Written Directly for the Screen': {u'nominees': [[u'Amour', u'1602620'], [u'Flight', u'1907668'], [u'Moonrise Kingdom', u'1748122'], [u'Zero Dark Thirty', u'1790885']], u'winner': [u'Django Unchained', u'1853728']}, u'Best Foreign Language Film': {u'nominees': [[u'War Witch', u'1820488'], [u'No', u'2059255'], [u'A Royal Affair', u'1276419'], [u'Kon-Tiki', u'1613750']], u'winner': [u'Amour', u'1602620']}, u'Best Film Editing': {u'nominees': [[u'Life of Pi', u'0454876'], [u'Lincoln', u'0443272'], [u'Silver Linings Playbook', u'1045658'], [u'Zero Dark Thirty', u'1790885']], u'winner': [u'Argo', u'1024648']}, u'Best Short Film, Live Action': {u'nominees': [[u'Asad', u'2136747'], [u'Buzkashi Boys', u'2133304'], [u'Death of a Shadow', u'2312702'], [u'Henry', u'2004244']], u'winner': [u'Curfew', u'2088735']}, u'Best Music, Original Song': {u'nominees': [[u'Ted', u'1637725'], [u'Life of Pi', u'0454876'], [u'Les Miserables', u'1707386'], [u'Chasing Ice', u'1579361']], u'winner': [u'Skyfall', u'1074638']}, u'Best Music, Original Score': {u'nominees': [[u'Anna Karenina', u'1781769'], [u'Argo', u'1024648'], [u'Lincoln', u'0443272'], [u'Skyfall', u'1074638']], u'winner': [u'Life of Pi', u'0454876']}, u'Best Production Design': {u'nominees': [[u'Anna Karenina', u'1781769'], [u'The Hobbit: An Unexpected Journey', u'0903624'], [u'Les Miserables', u'1707386'], [u'Life of Pi', u'0454876']], u'winner': [u'Lincoln', u'0443272']}, u'Best Short Film, Animated': {u'nominees': [[u'Adam and Dog', u'2162565'], [u'Fresh Guacamole', u'2309977'], [u'Head Over Heels', u'2391009'], [u'The Longest Daycare', u'2175842']], u'winner': [u'Paperman', u'2388725']}, u'Best Makeup': {u'nominees': [[u'Hitchcock', u'0975645'], [u'The Hobbit: An Unexpected Journey', u'0903624']], u'winner': [u'Les Miserables', u'1707386']}, u'Best Costume Design': {u'nominees': [[u'Les Miserables', u'1707386'], [u'Lincoln', u'0443272'], [u'Mirror Mirror', u'1667353'], [u'Snow White and the Huntsman', u'1735898']], u'winner': [u'Anna Karenina', u'1781769']}, u'Best Animated Feature': {u'nominees': [[u'Frankenweenie', u'1142977'], [u'ParaNorman', u'1623288'], [u'The Pirates! Band of Misfits', u'1430626'], [u'Wreck-It Ralph', u'1772341']], u'winner': [u'Brave', u'1217209']}, u'Best Writing, Screenplay Based on Material from Another Medium': {u'nominees': [[u'Beasts of the Southern Wild', u'2125435'], [u'Life of Pi', u'0454876'], [u'Lincoln', u'0443272'], [u'Silver Linings Playbook', u'1045658']], u'winner': [u'Argo', u'1024648']}, u'Best Sound Editing': {u'nominees': [[u'Zero Dark Thirty', u'1790885'], [u'Argo', u'1024648'], [u'Django Unchained', u'1853728'], [u'Life of Pi', u'0454876']], u'winner': [u'Skyfall', u'1074638']}, u'Best Actress in a Supporting Role': {u'nominees': [[u'The Master', u'1560747'], [u'Lincoln', u'0443272'], [u'The Sessions', u'1866249'], [u'Silver Linings Playbook', u'1045658']], u'winner': [u'Les Miserables', u'1707386']}, u'Best Cinematography': {u'nominees': [[u'Anna Karenina', u'1781769'], [u'Django Unchained', u'1853728'], [u'Lincoln', u'0443272'], [u'Skyfall', u'1074638']], u'winner': [u'Life of Pi', u'0454876']}, u'Best Documentary, Short Subjects': {u'nominees': [[u'Kings Point', u'2109153'], [u'Mondays at Racine', u'2083264'], [u'Open Heart', u'2348322'], [u'Redemption', u'2201886']], u'winner': [u'Inocente', u'2123210']}, u'Best Sound': {u'nominees': [[u'Argo', u'1024648'], [u'Life of Pi', u'0454876'], [u'Lincoln', u'0443272'], [u'Skyfall', u'1074638']], u'winner': [u'Les Miserables', u'1707386']}, u'Best Director': {u'nominees': [[u'Amour', u'1602620'], [u'Silver Linings Playbook', u'1045658'], [u'Lincoln', u'0443272'], [u'Beasts of the Southern Wild', u'2125435']], u'winner': [u'Life of Pi', u'0454876']}, u'Best Actor in a Leading Role': {u'nominees': [[u'Silver Linings Playbook', u'1045658'], [u'Les Miserables', u'1707386'], [u'The Master', u'1560747'], [u'Flight', u'1907668']], u'winner': [u'Lincoln', u'0443272']}, u'Best Picture': {u'nominees': [[u'Amour', u'1602620'], [u'Beasts of the Southern Wild', u'2125435'], [u'Django Unchained', u'1853728'], [u'Les Miserables', u'1707386'], [u'Life of Pi', u'0454876'], [u'Lincoln', u'0443272'], [u'Silver Linings Playbook', u'1045658'], [u'Zero Dark Thirty', u'1790885']], u'winner': [u'Argo', u'1024648']}, u'Best Actress in a Leading Role': {u'nominees': [[u'Zero Dark Thirty', u'1790885'], [u'Amour', u'1602620'], [u'Beasts of the Southern Wild', u'2125435'], [u'The Impossible', u'1649419']], u'winner': [u'Silver Linings Playbook', u'1045658']}, u'Best Effects, Visual Effects': {u'nominees': [[u'The Avengers', u'0848228'], [u'The Hobbit: An Unexpected Journey', u'0903624'], [u'Prometheus', u'1446714'], [u'Snow White and the Huntsman', u'1735898']], u'winner': [u'Life of Pi', u'0454876']}}, u'2012': {u'Best Actor in a Supporting Role': {u'nominees': [[u'My Week with Marilyn', u'1655420'], [u'Moneyball', u'1210166'], [u'Warrior', u'1291584'], [u'Extremely Loud & Incredibly Close', u'0477302']], u'winner': [u'Beginners', u'1532503']}, u'Best Documentary, Features': {u'nominees': [[u'Hell and Back Again', u'1748043'], [u'If a Tree Falls: A Story of the Earth Liberation Front', u'1787725'], [u'Paradise Lost 3: Purgatory', u'2028530'], [u'Pina', u'1440266']], u'winner': [u'Undefeated', u'1860355']}, u'Best Writing, Screenplay Written Directly for the Screen': {u'nominees': [[u'The Artist', u'1655442'], [u'Bridesmaids', u'1478338'], [u'Margin Call', u'1615147'], [u'A Separation', u'1832382']], u'winner': [u'Midnight in Paris', u'1605783']}, u'Best Foreign Language Film': {u'nominees': [[u'Bullhead', u'1821593'], [u'Footnote', u'1445520'], [u'In Darkness', u'1417075'], [u'Monsieur Lazhar', u'2011971']], u'winner': [u'A Separation', u'1832382']}, u'Best Film Editing': {u'nominees': [[u'The Artist', u'1655442'], [u'The Descendants', u'1033575'], [u'Hugo', u'0970179'], [u'Moneyball', u'1210166']], u'winner': [u'The Girl with the Dragon Tattoo', u'1568346']}, u'Best Art Direction': {u'nominees': [[u'War Horse', u'1568911'], [u'Midnight in Paris', u'1605783'], [u'Harry Potter and the Deathly Hallows: Part 2', u'1201607'], [u'The Artist', u'1655442']], u'winner': [u'Hugo', u'0970179']}, u'Best Short Film, Live Action': {u'nominees': [[u'Pentecost', u'1866218'], [u'Raju', u'1819651'], [u'Time Freak', u'1733689'], [u'Tuba Atlantic', u'2061843']], u'winner': [u'The Shore', u'2201259']}, u'Best Music, Original Song': {u'nominees': [[u'Rio', u'1436562']], u'winner': [u'The Muppets', u'1204342']}, u'Best Music, Original Score': {u'nominees': [[u'The Adventures of Tintin', u'0983193'], [u'Hugo', u'0970179'], [u'Tinker Tailor Soldier Spy', u'1340800'], [u'War Horse', u'1568911']], u'winner': [u'The Artist', u'1655442']}, u'Best Short Film, Animated': {u'nominees': [[u'Sunday', u'1895334'], [u'La luna', u'1957945'], [u'A Morning Stroll', u'1964446'], [u'Wild Life', u'2097943']], u'winner': [u'The Fantastic Flying Books of Mr. Morris Lessmore', u'1778342']}, u'Best Makeup': {u'nominees': [[u'Harry Potter and the Deathly Hallows: Part 2', u'1201607'], [u'Albert Nobbs', u'1602098']], u'winner': [u'The Iron Lady', u'1007029']}, u'Best Costume Design': {u'nominees': [[u'Anonymous', u'1521197'], [u'Hugo', u'0970179'], [u'Jane Eyre', u'1229822'], [u'W.E.', u'1536048']], u'winner': [u'The Artist', u'1655442']}, u'Best Writing, Screenplay Based on Material from Another Medium': {u'nominees': [[u'Hugo', u'0970179'], [u'The Ides of March', u'1124035'], [u'Tinker Tailor Soldier Spy', u'1340800'], [u'Moneyball', u'1210166']], u'winner': [u'The Descendants', u'1033575']}, u'Best Sound Editing': {u'nominees': [[u'The Girl with the Dragon Tattoo', u'1568346'], [u'Transformers: Dark of the Moon', u'1399103'], [u'War Horse', u'1568911'], [u'Drive', u'0780504']], u'winner': [u'Hugo', u'0970179']}, u'Best Actress in a Supporting Role': {u'nominees': [[u'The Artist', u'1655442'], [u'The Help', u'1454029'], [u'Bridesmaids', u'1478338'], [u'Albert Nobbs', u'1602098']], u'winner': [u'The Help', u'1454029']}, u'Best Cinematography': {u'nominees': [[u'The Artist', u'1655442'], [u'The Girl with the Dragon Tattoo', u'1568346'], [u'The Tree of Life', u'0478304'], [u'War Horse', u'1568911']], u'winner': [u'Hugo', u'0970179']}, u'Best Documentary, Short Subjects': {u'nominees': [[u'God Is the Bigger Elvis', u'2210633'], [u'The Barber of Birmingham: Foot Soldier of the Civil Rights Movement', u'1844056'], [u'Incident in New Baghdad', u'1746180'], [u'The Tsunami and the Cherry Blossom', u'2028578']], u'winner': [u'Saving Face', u'2140371']}, u'Best Sound': {u'nominees': [[u'Moneyball', u'1210166'], [u'War Horse', u'1568911'], [u'Transformers: Dark of the Moon', u'1399103'], [u'The Girl with the Dragon Tattoo', u'1568346']], u'winner': [u'Hugo', u'0970179']}, u'Best Director': {u'nominees': [[u'Midnight in Paris', u'1605783'], [u'The Tree of Life', u'0478304'], [u'The Descendants', u'1033575'], [u'Hugo', u'0970179']], u'winner': [u'The Artist', u'1655442']}, u'Best Actor in a Leading Role': {u'nominees': [[u'A Better Life', u'1554091'], [u'The Descendants', u'1033575'], [u'Tinker Tailor Soldier Spy', u'1340800'], [u'Moneyball', u'1210166']], u'winner': [u'The Artist', u'1655442']}, u'Best Picture': {u'nominees': [[u'The Tree of Life', u'0478304'], [u'The Descendants', u'1033575'], [u'Extremely Loud & Incredibly Close', u'0477302'], [u'The Help', u'1454029'], [u'Hugo', u'0970179'], [u'Midnight in Paris', u'1605783'], [u'Moneyball', u'1210166'], [u'War Horse', u'1568911']], u'winner': [u'The Artist', u'1655442']}, u'Best Actress in a Leading Role': {u'nominees': [[u'Albert Nobbs', u'1602098'], [u'The Help', u'1454029'], [u'The Girl with the Dragon Tattoo', u'1568346'], [u'My Week with Marilyn', u'1655420']], u'winner': [u'The Iron Lady', u'1007029']}, u'Best Effects, Visual Effects': {u'nominees': [[u'Real Steel', u'0433035'], [u'Harry Potter and the Deathly Hallows: Part 2', u'1201607'], [u'Transformers: Dark of the Moon', u'1399103'], [u'Rise of the Planet of the Apes', u'1318514']], u'winner': [u'Hugo', u'0970179']}, u'Best Animated Feature Film of the Year': {u'nominees': [[u'Chico & Rita', u'1235830'], [u'A Cat in Paris', u'1673702'], [u'Puss in Boots', u'0448694'], [u'Kung Fu Panda 2', u'1302011']], u'winner': [u'Rango', u'1192628']}}}\n"
]
}
],
"prompt_number": 92
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"However, the names of some of the different awards have changed over the years; we had better fix this for consistency later on. "
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"movie_awards = [u'Best Documentary, Features',u'Best Writing, Screenplay Written Directly for the Screen',\n",
" u'Best Foreign Language Film',u'Best Short Film, Live Action',u'Best Music, Original Song',\n",
" u'Best Music, Original Score',u'Best Art Direction-Set Decoration',u'Best Costume Design',\n",
" u'Best Short Film, Animated',u'Best Makeup',u'Best Film Editing',u'Best Animated Feature',\n",
" u'Best Writing, Screenplay Based on Material from Another Medium',u'Best Sound Editing',\n",
" u'Best Cinematography',u'Best Documentary, Short Subjects',u'Best Sound',u'Best Director',\n",
" u'Best Picture',u'Best Visual Effects']\n",
"\n",
"# http://stackoverflow.com/questions/9708374/python-how-to-find-the-string-with-most-matches-in-a-list-of-strings\n",
"def find_correct_award(awards,keystr):\n",
" keywords = re.split(' |-',keystr.lower())\n",
" def matches(text):\n",
" return sum(word in text.lower() for word in keywords)\n",
" return max(awards,key=matches) \n",
"\n",
"for year in oscars:\n",
" for award in movie_awards:\n",
" if award not in oscars[year].keys():\n",
" old = find_correct_award(oscars[year].keys(),award)\n",
" oscars[year][award] = oscars[year].pop(old)"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 109
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"print oscars['2003']['Best Cinematography']"
],
"language": "python",
"metadata": {},
"outputs": [
{
"output_type": "stream",
"stream": "stdout",
"text": [
"{u'nominees': [[u'Chicago', u'0299658'], [u'Far from Heaven', u'0297884'], [u'Gangs of New York', u'0217505'], [u'The Pianist', u'0253474']], u'winner': [u'Road to Perdition', u'0257044']}\n"
]
}
],
"prompt_number": 110
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Populating a List of Candidate Movies
"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"In order to run our analysis, we need to isolate a dataset of movies for a given year that are candidates for an Oscar. For this purpose, we decided to extract a subset of the movies outlined by Box Office Mojo as the top movies in a given year. Assumming that the movies most likely to win an Oscar are likely to be among those with the highest gross revenue in a given year, we decided to select the Top 200 movies from each year as candidates for an Oscar (in the following year)."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Getting the Movies
"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"The function get_top_movies_by_year takes two arguments, year and the number of movies desired for that given year. The function returns a dataframe for the given year with the following columns:\n",
"\n",
"- rank
- year
- movie_title
- imdbid
- rtid
- studio
- total_gross
- num_theaters_total
- opening_revenue
- num_theaters_opening
- date_open
- date_close
\n",
"
\n",
"This data, with the exception of IMDBid and RTid, is obtained from Box Office Mojo. In a first pass, we attempt to collect IMDBid and RTid from Rotten Tomatoes API. For the movies that we cannot successfully isolate an IMDBid for, we attempt to obtain an id through the OMDB API."
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"def get_top_movies_by_year(yr,num_movies=200):\n",
" num_pages = int((num_movies-1)/100) + 1\n",
" missing_rtid = 0\n",
" missing_imdbid = 0\n",
" alldata = []\n",
" for pagenum in range(1,num_pages+1):\n",
" url = 'http://boxofficemojo.com/yearly/chart/?page=' + str(pagenum) + '&view=releasedate&view2=domestic&yr=' + str(yr) + '&p=.htm'\n",
" page_xml = requests.get(url).text\n",
" dom = web.Element(page_xml)\n",
" for table in dom.by_tag('table'):\n",
" cellpadding = (table.attributes['cellpadding']) \n",
" if (cellpadding == '5'): # isolates table with movie data\n",
" for row in table.by_tag('tr'):\n",
" cols = row.by_tag('td')\n",
" #catches movie rows\n",
" try:\n",
" rank = int(web.strip_tags(cols[0]))\n",
" if (rank<=num_movies): #checks to make sure in right spot!\n",
" movie_title = web.strip_tags(cols[1]).strip('\\t')\n",
" movie_title = unicodedata.normalize('NFKD', movie_title).encode('ascii', 'ignore') #removes accents\n",
" studio = web.strip_tags(cols[2]).strip('\\t')\n",
" \n",
" # the below try/except mechanisms catch errors when there is a NONE instead of a string which is a candidate for integer casting\n",
" try:\n",
" total_gross = int(((web.strip_tags(cols[3])).strip('$')).replace(',', ''))\n",
" except:\n",
" total_gross = None\n",
" try:\n",
" num_theaters_total = int((web.strip_tags(cols[4])).replace(',', ''))\n",
" except:\n",
" num_theaters_total = None\n",
" try:\n",
" opening_revenue = int(((web.strip_tags(cols[5])).strip('$')).replace(',', ''))\n",
" except:\n",
" opening_revenue = None\n",
" try: \n",
" num_theaters_opening = int((web.strip_tags(cols[6])).replace(',', ''))\n",
" except:\n",
" num_theaters_opening = None\n",
" \n",
" date_open = web.strip_tags(cols[7]).strip('\\t')\n",
" date_close = web.strip_tags(cols[8]).strip('\\t')\n",
" \n",
" #using rotten tomatoes to get rtid/imdbid\n",
" url2 = 'http://api.rottentomatoes.com/api/public/v1.0/movies.json?' 'q=' + movie_title + ' &page_limit=10' + '&apikey=' + api_key\n",
" info_data = requests.get(url2).text\n",
" info_json = json.loads(info_data)\n",
" movie_count = 0\n",
" rtid=''\n",
" imdbid=''\n",
" try:\n",
" for movie in info_json['movies']:\n",
" if movie['year']==yr:\n",
" movie_count+=1\n",
" rtid=movie['id']\n",
" try:\n",
" imdbid=(movie['alternate_ids'])['imdb']\n",
" except:\n",
" imdbid=''\n",
" except:\n",
" pass\n",
" if movie_count!=1: #case where more than one movie met description\n",
" rtid = ''\n",
" imdbid = ''\n",
" \n",
" # second attempt to obtain imdb ids using the omdbapi\n",
" if imdbid=='':\n",
" try:\n",
" omdb_link = \"http://www.omdbapi.com/?t=%s&y=%s\" % (movie_title,yr)\n",
" omdb_text = web.URL(omdb_link).download(cached=True)\n",
" imdbid = json.loads(omdb_text)['imdbID'].replace(\"tt\",\"\")\n",
" except:\n",
" try:\n",
" omdb_link = \"http://www.omdbapi.com/?t=%s&y=%s\" % (movie_title,(yr-1))\n",
" omdb_text = web.URL(omdb_link).download(cached=True)\n",
" imdbid = json.loads(omdb_text)['imdbID'].replace(\"tt\",\"\")\n",
" except:\n",
" try:\n",
" omdb_link = \"http://www.omdbapi.com/?t=%s&y=%s\" % (movie_title,(yr-2))\n",
" omdb_text = web.URL(omdb_link).download(cached=True)\n",
" imdbid = json.loads(omdb_text)['imdbID'].replace(\"tt\",\"\")\n",
" except:\n",
" imdbid = ''\n",
" missing_imdbid += 1\n",
" if rtid=='':\n",
" missing_rtid += 1\n",
" \n",
" data_row = [rank,yr,movie_title,imdbid,rtid,studio,total_gross,num_theaters_total,opening_revenue,num_theaters_opening,date_open,date_close]\n",
" alldata.append(data_row)\n",
" except ValueError:\n",
" pass\n",
" \n",
" result = pd.DataFrame(alldata,columns=['rank','year','movie_title','imdbid','rtid','studio','total_gross','num_theaters_total','opening_revenue','num_theaters_opening','date_open','date_close'])\n",
" if (missing_imdbid>0):\n",
" print 'unable to find a imdbid for ' + str(missing_imdbid) + ' movies of ' + str(num_movies) + ' in ' + str(yr)\n",
" if (missing_rtid>0):\n",
" print 'unable to find a rtid for ' + str(missing_rtid) + ' movies of ' + str(num_movies) + ' in ' + str(yr)\n",
" return result"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 111
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"In order to obtain the candidate movies for the timerange of the data we are investigating (2003-2012), we must run get_top_movies_by_year and create a unified dataframe of movie data. The datafile 'alldata_200.csv' includes all of the data outlined previously that we obtain from Box Office Mojo. The datafile 'iddata_200.csv' only includes the data needed by portions of the project in order to collect additional movie data, namely: year, movie rank, imdbid, and rtid."
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"# Saves all data\n",
"# alldata=get_top_movies_by_year(2003,200)\n",
"# for yr in range(2004,2013):\n",
"# num_movies=200\n",
"# result=get_top_movies_by_year(yr,num_movies)\n",
"# alldata.append(result,ignore_index=True)\n",
"# print yr\n",
"# filename='./data/alldata_' + str(num_movies) + '.csv'\n",
"# alldata.to_csv(filename, index=False)\n",
"\n",
"# Saves just title and id data in a separate file\n",
"# narrowed_data=pd.DataFrame(alldata, columns=['year','rank','movie_title','imdbid','rtid'])\n",
"# filename='./data/iddata_' + str(num_movies) + '.csv'\n",
"# narrowed_data.to_csv(filename, index=False)"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 112
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Data Refinement and Cleaning
\n",
"The data collected via get_top_movies by year successfully found the IMDBid (the reference id used across the different segment of this project) for approximately 90% of the top 200 movies in each given year. The additional 10% of IMDBids not found by the above procedure were identified manually. The manual determination of IMDBids was recorded in two new files: 'revised_id_data.csv' and 'completedata.csv'.\n",
"\n",
"Of the top 200 movies from each year, approximately 5% were excluded from further analysis. The top reasons for movie exclusion included:\n",
"\n",
"- Re-releases of prior movies (often occurs around the holidays)
\n",
"- Re-releases of movies with visual enhancements (a very common procedure for Disney movies)
\n",
"- Exclusive release of movies in IMAX theaters
\n",
"- One-time limited showings of events such as concerts (see Fathom Events)
\n",
"
\n",
"\n",
"When a movie was excluded from analysis its IMDBid was set to 'None'. These rows of the dataframe were then removed from analysis."
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"bomj_df = pd.read_csv('data/completedata.csv')\n",
"\n",
"def test_for_nones(df,column_name):\n",
" result = []\n",
" col = df[column_name]\n",
" for item in col:\n",
" check_bool = False\n",
" if (item != 'None'):\n",
" if item!='NaN':\n",
" check_bool = True\n",
" result.append(check_bool)\n",
" return result\n",
"\n",
"# stores imdbid with 7 characters\n",
"bomj_df=bomj_df[test_for_nones(bomj_df,'imdbid')]\n",
"# removes rtid, which is calculated later\n",
"bomj_df=bomj_df[['rank','year','movie_title','imdbid','studio','total_gross','num_theaters_total','opening_revenue','num_theaters_opening','date_open','date_close']]\n",
"\n",
"import numpy as np\n",
"\n",
"# ensures proper formatting of all columns\n",
"def formatting_verification(column,fun):\n",
" result = []\n",
" for val in column:\n",
" if fun=='int':\n",
" if ((val == 'None')):\n",
" result.append(None)\n",
" else:\n",
" print val\n",
" result.append(int(val))\n",
"\n",
"# elif fun=='str':\n",
"# if ((val is None)):\n",
"# result.append()\n",
"# else:\n",
"# try:\n",
"# if math.isnan(val):\n",
"# result.append()\n",
"# else:\n",
"# result.append(str(val))\n",
"# except:\n",
"# result.append(str(val))\n",
"# return result \n",
"\n",
"\n",
"bomj_df['imdbid'] = bomj_df['imdbid'].map(lambda id: str(id).zfill(7))\n",
"\n",
"# bomj_df['rank'] = formatting_verification(bomj_df['rank'].values,'int')\n",
"# bomj_df['year'] = formatting_verification(bomj_df['year'].values,'int')\n",
"# bomj_df['studio'] = formatting_verification(bomj_df['studio'].values,'str')\n",
"# bomj_df['total_gross'] = formatting_verification(bomj_df['total_gross'].values,'int')\n",
"#bomj_df['num_theaters_total'] = formatting_verification(bomj_df['num_theaters_total'].values,'int')\n",
"# bomj_df['opening_revenue'] = formatting_verification(bomj_df['opening_revenue'].values,'int')\n",
"# bomj_df['num_theaters_opening'] = formatting_verification(bomj_df['num_theaters_opening'].values,'int')\n",
"# bomj_df['date_open'] = pd.to_datetime(bomj_df['date_open'])\n",
"# bomj_df['date_close'] = pd.to_datetime(bomj_df['date_close'])\n",
"#bomj_df[['num_theaters_total']] = bomj_df[['num_theaters_total']].astype(int)\n",
"#bomj_df['num_theaters_total'] = formatting_verification(bomj_df['num_theaters_total'].values,'int')"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 163
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Verifying Movie Selection
"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"imdb_as_index = bomj_df.set_index('imdbid')\n",
"imdb_as_index.head()\n",
"data_by_year = imdb_as_index.groupby('year').groups\n",
"performance = {}\n",
"total_nominees_present = 0\n",
"total_nominees_absent = 0\n",
"total_winners_present = 0\n",
"total_winners_absent = 0\n",
"for (yr,yr_list) in oscars.iteritems():\n",
" yr = int(yr)\n",
" if yr>2003:\n",
" yr=yr-1\n",
" movieids=data_by_year[yr]\n",
" nominee_ids=[]\n",
" winner_ids=[]\n",
" \n",
" nominees_present = 0\n",
" nominees_absent = 0\n",
" winners_present = 0\n",
" winners_absent = 0\n",
" \n",
" for (award,award_list) in yr_list.iteritems():\n",
" winner_ids.append(award_list['winner'][1])\n",
" for nominee in award_list['nominees']:\n",
" nominee_ids.append(nominee[1])\n",
" for nid in nominee_ids:\n",
" is_movie_saved = False\n",
" for top200id in movieids:\n",
" try:\n",
" corrected_id=str(int(top200id)).zfill(7)\n",
" if corrected_id==nid:\n",
" is_movie_saved = True\n",
" except:\n",
" pass\n",
" if is_movie_saved:\n",
" nominees_present += 1\n",
" else:\n",
" nominees_absent += 1\n",
" for wid in winner_ids:\n",
" is_movie_saved = False\n",
" for top200id in movieids:\n",
" try:\n",
" corrected_id=str(int(top200id)).zfill(7)\n",
" if corrected_id==wid:\n",
" is_movie_saved = True\n",
" except:\n",
" pass\n",
" if is_movie_saved:\n",
" winners_present += 1\n",
" else:\n",
" winners_absent += 1 \n",
" total_nominees_present += nominees_present\n",
" total_nominees_absent += nominees_absent\n",
" total_winners_present += winners_present\n",
" total_winners_absent += winners_absent\n",
" performance[yr]=[winners_present,winners_absent,nominees_present,nominees_absent]\n",
" #print str(yr) + ' nominee performance: ' + str(float(nominees_present)/(nominees_present+nominees_absent))\n",
" #print str(yr) + ' winner performance: ' + str(float(winners_present)/(winners_present+winners_absent))\n",
" #print\n",
"\n",
"print 'total nominee performance: ' + str(float(total_nominees_present)/(total_nominees_present+total_nominees_absent))\n",
"print 'total winner performance: ' + str(float(total_winners_present)/(total_winners_present+total_winners_absent))"
],
"language": "python",
"metadata": {},
"outputs": [
{
"output_type": "stream",
"stream": "stdout",
"text": [
"total nominee performance: 0.769230769231\n",
"total winner performance: 0.7875\n"
]
}
],
"prompt_number": 20
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Getting the Reviews
\n",
"\n",
"We need to constuct dataframes of reviews of movies from different review sources. We will be using the following sources:\n",
"\n",
"- Rotten Tomatoes
\n",
"- Metacritic
\n",
"- IMDB
\n",
"
\n",
"\n",
"For each of these, we will construct a separate dataframe with the following structure:\n",
"\n",
" \n",
" \n",
" | Critic | \n",
" Normalized Score | \n",
" Quote | \n",
" ID | \n",
" Title | \n",
" Source | \n",
" Overall Score | \n",
" Year | \n",
"
\n",
" \n",
" \n",
" \n",
" | Name of the Critic | \n",
" Score normalized to what the max is for this source | \n",
" The string of the review | \n",
" Movie ID for this site | \n",
" Title of the movie | \n",
" Source of the review | \n",
" The overall score for this movie on this site (if there is one) | \n",
" Year that the movie is form | \n",
"
\n",
" \n",
"
"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"print \"For Example:\"\n",
"pd.DataFrame({\"critic\":[''],\"norm_score\":[''],\"quote\":[''],\"id\":[''],\"title\":[''],\"source\":[''],\"overall_score\":[''],\"year\":[''],\"date\":['']}).head()"
],
"language": "python",
"metadata": {},
"outputs": [
{
"output_type": "stream",
"stream": "stdout",
"text": [
"For Example:\n"
]
},
{
"html": [
"\n",
"
\n",
" \n",
" \n",
" | \n",
" critic | \n",
" date | \n",
" id | \n",
" norm_score | \n",
" overall_score | \n",
" quote | \n",
" source | \n",
" title | \n",
" year | \n",
"
\n",
" \n",
" \n",
" \n",
" | 0 | \n",
" | \n",
" | \n",
" | \n",
" | \n",
" | \n",
" | \n",
" | \n",
" | \n",
" | \n",
"
\n",
" \n",
"
\n",
"
Rotten Tomatoes
"
]
},
{
"cell_type": "code",
"collapsed": true,
"input": [
"# write the functions to scrape the data\n",
"\n",
"'''\n",
"Get Individual Review From Rotten Tomatoes\n",
"\n",
"'''\n",
"\n",
"def get_rtomatoes_movie_reviews(imdb):\n",
"\t#api_key = 't97y5fkhfbfu9wjhse7v8ayy' \n",
"\t#api_key = '8kr5nqyrshjyp4y69w8rzzgu'\n",
"\tapi_key = 'sdat77wpkdnuyvb9bqhz2y5v'\n",
"\tif(len(imdb)<7):\n",
"\t\textra = 7 - len(imdb)\n",
"\t\tprefix = \"\"\n",
"\t\tfor i in range(0, extra):\n",
"\t\t\tprefix = prefix + '0'\n",
"\t\t\timdb = prefix + imdb\n",
"\n",
"\turl = 'http://api.rottentomatoes.com/api/public/v1.0/movie_alias.json' \n",
"\toptions = {'type': 'imdb','id': imdb, 'apikey': api_key}\n",
"\trotIds = requests.get(url, params=options).text\n",
"\trotIds = json.loads(rotIds)\n",
"\n",
"\tif('\"error\":\"Could not find a movie with the specified id\"' in rotIds):\n",
"\t\treturn None\n",
"\tif('id' not in rotIds):\n",
"\t\treturn None\n",
"\n",
"\ttitle = rotIds['title']\n",
"\tyear = rotIds['year']\n",
"\toverall_score = rotIds['ratings']['critics_score']\n",
"\trotId = rotIds['id']\n",
"\t\n",
"\turl2 = 'http://api.rottentomatoes.com/api/public/v1.0/movies/%s/reviews.json' % rotId\n",
"\toptions2 = {'review_type': 'top_critic', 'page_limit': 50, 'page': 1, 'apikey': api_key}\n",
"\tdata = requests.get(url2, params=options2).text\n",
"\tdata = json.loads(data) # load a json string into a collection of lists and dicts\n",
"\tif('total' not in data or 'reviews' not in data):\n",
"\t\treturn None\n",
"\tif str(data['total']) == '0':\n",
"\t\treturn None\n",
"\n",
"\tids = []\n",
"\ttitles = []\n",
"\tyears = []\n",
"\toverall_scores = []\n",
"\t\n",
"\trevs = data['reviews'] \n",
"\tcritics = []\n",
"\tquotes = []\n",
"\tpubs = []\n",
"\tfreshness = []\n",
"\tdates = []\n",
"\n",
"\tfor each in revs:\n",
"\t\tids.append(imdb)\n",
"\t\ttitles.append(title)\n",
"\t\tyears.append(year)\n",
"\t\toverall_scores.append(float(overall_score)/100.0)\n",
"\n",
"\t\tcritics.append(each['critic'])\n",
"\t\tquotes.append(each['quote'])\n",
"\t\tpubs.append(each['publication'])\n",
"\t\tif(each['freshness'] == 'fresh'):\n",
"\t\t\tfreshness.append(1)\n",
"\t\telse:\n",
"\t\t\tfreshness.append(0)\n",
"\t\tdates.append(each['date']) \n",
"\t\n",
"\td = {'critic' : critics, 'norm_score' : freshness, 'quote' : quotes, 'id' : ids, 'title' : titles, 'source' : pubs, 'overall_score' : overall_scores, 'date' : dates, 'year': years} \n",
"\treturn pd.DataFrame(d)\n",
"\n",
"def get_all_rtomatoes_movies(movieList):\n",
"\tall_imdbs = alldata['imdbid']\n",
"\timdbs = []\n",
"\tfor each in all_imdbs:\n",
"\t\tif str(each) != 'None':\n",
"\t\t\timdbs.append(each)\n",
"\t#print imdbs\n",
"\t#for each in imdbs:\n",
"\n",
"\tnewDF = pd.DataFrame()\n",
"\tfor each in imdbs:\n",
"\t\tindividualDF = get_rtomatoes_movie_reviews(each)\n",
"\t\tif(individualDF):\n",
"\t\t\tnewDF = newDF.append(individualDF, ignore_index=True)\n",
"\n",
"\n",
"\tfilename='./data/rotten_tomatoes_data.csv'\n",
"\tnewDF.to_csv(filename, index=False)\n",
"\treturn newDF"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 3
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"# actually call the functions and save in a file\n",
"\n",
"#alldata = pd.read_csv('./revised_id_data.csv') \n",
"#rot_tom_data = get_all_rtomatoes_movies(alldata)\n",
"\n",
"#once we have the stored csv file not necessary to make the above calls, and can just do:\n",
"rotten_df = pd.read_csv('data/rotten_tomatoes_data.csv')\n",
"\n",
"def filter_tomatoes(dataF):\n",
" for ind, each in enumerate(dataF['quote']):\n",
" if(str(each) == 'nan'):\n",
" dataF['quote'][ind] = 'NoReview'\n",
" dataF = dataF[dataF.quote != 'NoReview']\n",
" return dataF\n",
"\n",
"rotten_df = filter_tomatoes(rotten_df)\n",
"rotten_df.head()\n"
],
"language": "python",
"metadata": {},
"outputs": [
{
"html": [
"\n",
"
\n",
" \n",
" \n",
" | \n",
" critic | \n",
" date | \n",
" id | \n",
" norm_score | \n",
" overall_score | \n",
" quote | \n",
" source | \n",
" title | \n",
" year | \n",
"
\n",
" \n",
" \n",
" \n",
" | 0 | \n",
" David Germain | \n",
" 11/19/13 | \n",
" 167260 | \n",
" 1 | \n",
" 0.94 | \n",
" With The Lord of the Rings: The Return of the ... | \n",
" Associated Press | \n",
" The Lord of the Rings: The Return of the King | \n",
" 2003 | \n",
"
\n",
" \n",
" | 1 | \n",
" Joe Morgenstern | \n",
" 11/19/13 | \n",
" 167260 | \n",
" 1 | \n",
" 0.94 | \n",
" Yes, the running time is long, and yes, those ... | \n",
" Wall Street Journal | \n",
" The Lord of the Rings: The Return of the King | \n",
" 2003 | \n",
"
\n",
" \n",
" | 2 | \n",
" Andrea Gronvall | \n",
" 2/6/07 | \n",
" 167260 | \n",
" 1 | \n",
" 0.94 | \n",
" Standing out amid an excellent cast is Elijah ... | \n",
" Chicago Reader | \n",
" The Lord of the Rings: The Return of the King | \n",
" 2003 | \n",
"
\n",
" \n",
" | 3 | \n",
" Geoff Andrew | \n",
" 2/9/06 | \n",
" 167260 | \n",
" 0 | \n",
" 0.94 | \n",
" Some story strands are crudely abbreviated; ot... | \n",
" Time Out | \n",
" The Lord of the Rings: The Return of the King | \n",
" 2003 | \n",
"
\n",
" \n",
" | 4 | \n",
" Peter Travers | \n",
" 1/15/04 | \n",
" 167260 | \n",
" 1 | \n",
" 0.94 | \n",
" Pops your eyes, excites your senses and brings... | \n",
" Rolling Stone | \n",
" The Lord of the Rings: The Return of the King | \n",
" 2003 | \n",
"
\n",
" \n",
"
\n",
"
Metacritic
"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"# write the functions to scrape the data"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 13
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"# actually call the functions and save in a file"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 14
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"IMDB
"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"\"\"\"\n",
"Function\n",
"--------\n",
"get_imdb_movie_reviews:\n",
" retrieves the movie reviews for a given movie on imdb\n",
"Parameters\n",
"----------\n",
"id:\n",
" imdb movie ID\n",
"title:\n",
" title of the movie (why scrape it when we already have it?)\n",
"year:\n",
" year that the movie was made\n",
"Returns\n",
"-------\n",
"returns a dataframe containing all of the movie revies of the form\n",
"listed above\n",
"\"\"\"\n",
"def get_imdb_movie_reviews(id,title,year):\n",
" score_max = 10.0\n",
" link = \"http://www.imdb.com/title/tt%0.7d/\" % id\n",
" url = web.URL(link)\n",
" dom = web.DOM(url.download(cached=True))\n",
" overall = float(dom.by_class(\"titlePageSprite star-box-giga-star\")[0].content.strip()) / score_max\n",
" # try to get year directly from page; this isn't present in every entry\n",
" try:\n",
" year = dom('span.itemprop[itemprop=name]')[0].next.next.by_tag('a')[0].content\n",
" year = int(year)\n",
" except:\n",
" pass\n",
" rc = dom.by_attr(itemprop=\"reviewCount\")[0].content.split(\" \")[0].replace(\",\",\"\")\n",
" revlink = link + 'reviews?count=%s&start=0' % rc # get at most 20 reviews\n",
" url = web.URL(revlink)\n",
" dom = web.DOM(url.download(cached=True))\n",
" parser = HTMLParser()\n",
" lst = []\n",
" hrs = dom.by_id('tn15main').by_tag('hr')\n",
" for hr in hrs:\n",
" div = hr.next.next\n",
" try:\n",
" score = float(div.by_tag(\"img\")[1].attrs[\"alt\"].split(\"/\")[0]) / score_max\n",
" date = div.by_tag(\"small\")[2].content\n",
" except:\n",
" continue\n",
" user = div.by_tag(\"a\")[1].content\n",
" p = div.next.next\n",
" review = parser.unescape(p.content.replace(\"
\",\"\\n\"))\n",
" lst.append(dict(critic=user,norm_score=score,quote=review,\n",
" id=id,title=title,source=\"IMDB\",overall_score=overall,year=year,date=date))\n",
" return lst\n",
"\n",
"def get_all_imdb_movies():\n",
" fname = 'revised_id_data.csv'\n",
" df = []\n",
" result = pd.read_csv('data/%s' % fname)\n",
" f = open('errors.txt','w+')\n",
" year = 2003\n",
" for index,row in result.iterrows():\n",
" try: \n",
" row['imdbid'] = int(row['imdbid'])\n",
" except:\n",
" continue\n",
" newyr = row['year']\n",
" if(int(newyr) != year and df != []):\n",
" # filter_quotes(df)\n",
" fo = open('imdbrev%d.json' % year,'w+')\n",
" json.dump(df,fo)\n",
" fo.close()\n",
" year = int(newyr)\n",
" df = []\n",
" gc.collect()\n",
" try:\n",
" df.append(get_imdb_movie_reviews(row['imdbid'],row['movie_title'],row['year']))\n",
" print row['movie_title'], year\n",
" except:\n",
" print >> f, \"ERROR: %s\" % row['movie_title']\n",
" f.close()\n",
" if df:\n",
" fo = open('imdbrev%d.json' % year,'w+')\n",
" json.dump(df,fo)\n",
" fo.close()"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 15
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"# create the json files for all of the imdb reviews\n",
"# get_all_imdb_movies()"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 16
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Although we had all the data for imdb now, we still had to clean it. It turns out there was were a log of unicode characters in the quotes that exporting to csv doesn't really play nice with. In order to overcome this, we had to clean the data removing all escaped sequences including '\\r','\\n','\\t', and all unicode characters."
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"def filter_quotes(df):\n",
" for index,row in df.iterrows():\n",
" text = row['quote']\n",
" text = re.sub(r\"\\\\u[0-9a-zAZ]{4}\", \"\", text)\n",
" text = re.sub(\"\\\\\\[^\\'|\\\"]\", \"\", text)\n",
" text = text.replace('\\n','')\n",
" text = unicodedata.normalize('NFKC',unicode(text)).encode('ascii','ignore')\n",
" df.ix[index, 'quote'] = text"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 17
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"\"\"\"\n",
"Here we construct the dataframe for all of the imdb reviews.\n",
"We then filter it, and save it to a csv.\n",
"\"\"\"\n",
"\"\"\"\n",
"imdb_f_list = ['imdbrev2003.json', 'imdbrev2004.json', 'imdbrev2005.json', \n",
" 'imdbrev2006.json', 'imdbrev2007.json', 'imdbrev2008.json', \n",
" 'imdbrev2009.json', 'imdbrev2010.json', 'imdbrev2011.json', 'imdbrev2012.json'] \n",
"\n",
"\n",
"imdb_df = pd.DataFrame({\"critic\":[],\"norm_score\":[],\"quote\":[],\"id\":[],\n",
" \"title\":[],\"source\":[],\"overall_score\":[],\"year\":[],\"date\":[]})\n",
"\n",
"for f in imdb_f_list:\n",
" fp = open('data/' + f,'rb')\n",
" j = json.load(fp)\n",
" fp.close()\n",
" for m in j:\n",
" if m != []:\n",
" imdb_df = imdb_df.append(m)\n",
" gc.collect()\n",
" \n",
"filter_quotes(imdb_df)\n",
"imdb_df.to_csv('data/imdb_reviews.csv',index=False)\n",
"\"\"\"\n",
"pass"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 18
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"# read saved dataframe\n",
"imdb_df = pd.read_csv('data/imdb_reviews.csv')\n",
"imdb_df.head()"
],
"language": "python",
"metadata": {},
"outputs": [
{
"ename": "KeyboardInterrupt",
"evalue": "",
"output_type": "pyerr",
"traceback": [
"\u001b[0;31m---------------------------------------------------------------------------\u001b[0m\n\u001b[0;31mKeyboardInterrupt\u001b[0m Traceback (most recent call last)",
"\u001b[0;32m\u001b[0m in \u001b[0;36m\u001b[0;34m()\u001b[0m\n\u001b[1;32m 1\u001b[0m \u001b[0;31m# read saved dataframe\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 2\u001b[0;31m \u001b[0mimdb_df\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mpd\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mread_csv\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m'data/imdb_reviews.csv'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 3\u001b[0m \u001b[0mimdb_df\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mhead\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
"\u001b[0;32m//anaconda/python.app/Contents/lib/python2.7/site-packages/pandas/io/parsers.pyc\u001b[0m in \u001b[0;36mparser_f\u001b[0;34m(filepath_or_buffer, sep, dialect, compression, doublequote, escapechar, quotechar, quoting, skipinitialspace, lineterminator, header, index_col, names, prefix, skiprows, skipfooter, skip_footer, na_values, na_fvalues, true_values, false_values, delimiter, converters, dtype, usecols, engine, delim_whitespace, as_recarray, na_filter, compact_ints, use_unsigned, low_memory, buffer_lines, warn_bad_lines, error_bad_lines, keep_default_na, thousands, comment, decimal, parse_dates, keep_date_col, dayfirst, date_parser, memory_map, nrows, iterator, chunksize, verbose, encoding, squeeze, mangle_dupe_cols, tupleize_cols)\u001b[0m\n\u001b[1;32m 398\u001b[0m )\n\u001b[1;32m 399\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 400\u001b[0;31m \u001b[0;32mreturn\u001b[0m \u001b[0m_read\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mfilepath_or_buffer\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mkwds\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 401\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 402\u001b[0m \u001b[0mparser_f\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m__name__\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mname\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
"\u001b[0;32m//anaconda/python.app/Contents/lib/python2.7/site-packages/pandas/io/parsers.pyc\u001b[0m in \u001b[0;36m_read\u001b[0;34m(filepath_or_buffer, kwds)\u001b[0m\n\u001b[1;32m 203\u001b[0m \u001b[0;32mreturn\u001b[0m \u001b[0mparser\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 204\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 205\u001b[0;31m \u001b[0;32mreturn\u001b[0m \u001b[0mparser\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mread\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 206\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 207\u001b[0m _parser_defaults = {\n",
"\u001b[0;32m//anaconda/python.app/Contents/lib/python2.7/site-packages/pandas/io/parsers.pyc\u001b[0m in \u001b[0;36mread\u001b[0;34m(self, nrows)\u001b[0m\n\u001b[1;32m 606\u001b[0m \u001b[0;31m# self._engine.set_error_bad_lines(False)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 607\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 608\u001b[0;31m \u001b[0mret\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_engine\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mread\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mnrows\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 609\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 610\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0moptions\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mget\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m'as_recarray'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
"\u001b[0;32m//anaconda/python.app/Contents/lib/python2.7/site-packages/pandas/io/parsers.pyc\u001b[0m in \u001b[0;36mread\u001b[0;34m(self, nrows)\u001b[0m\n\u001b[1;32m 1026\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1027\u001b[0m \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1028\u001b[0;31m \u001b[0mdata\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_reader\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mread\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mnrows\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 1029\u001b[0m \u001b[0;32mexcept\u001b[0m \u001b[0mStopIteration\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1030\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mnrows\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0mNone\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
"\u001b[0;31mKeyboardInterrupt\u001b[0m: "
]
}
],
"prompt_number": 19
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Exploring the Data
\n",
"\n",
"Here we visualize the reviews from the different sources to get a sense of the number of review, distribution of ratings, ..."
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"# now it's time to add the data\n",
"\"\"\"\n",
"Function\n",
"--------\n",
"add_oscar_data:\n",
" Adds data for being an oscar winner or nominee in place in the passed dataframe\n",
"\n",
"Parameters\n",
"----------\n",
" df: Dataframe\n",
" award_name: Name of the oscar award\n",
" If None, then creates general columns for winning any or being nominated in any\n",
"\"\"\"\n",
"\n",
"movie_awards = [u'Best Documentary, Features',u'Best Writing, Screenplay Written Directly for the Screen',\n",
" u'Best Foreign Language Film',u'Best Short Film, Live Action',u'Best Music, Original Song',\n",
" u'Best Music, Original Score',u'Best Art Direction-Set Decoration',u'Best Costume Design',\n",
" u'Best Short Film, Animated',u'Best Makeup',u'Best Film Editing',u'Best Animated Feature',\n",
" u'Best Writing, Screenplay Based on Material from Another Medium',u'Best Sound Editing',\n",
" u'Best Cinematography',u'Best Documentary, Short Subjects',u'Best Sound',u'Best Director',\n",
" u'Best Picture',u'Best Visual Effects']\n",
"\n",
"def add_oscar_data(df,imdbid_title='id',award_name=None):\n",
" awards = movie_awards\n",
" if award_name:\n",
" awards = [award_name]\n",
" win = award_name.replace(' ','_') + '_winner'\n",
" nom = award_name.replace(' ','_') + '_nom'\n",
" else:\n",
" win = 'oscar_winner'\n",
" nom = 'oscar_nom'\n",
" df[win] = 0\n",
" df[nom] = 0\n",
" for year in oscars:\n",
" for award in movie_awards:\n",
" try:\n",
" df[win][df[imdbid_title] == oscars[year][award]['winner'][1]] = 1\n",
" df[nom][df[imdbid_title] == oscars[year][award]['winner'][1]] = 1\n",
" for n in oscars[year][award]['nominees']:\n",
" df[nom][df[imdbid_title] == n[1]] = 1\n",
" # skip awards that don't exist for a year\n",
" except:\n",
" pass"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 126
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Box Office Mojo Data
"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"add_oscar_data(bomj_df,'imdbid')\n",
"bomj_df.head()"
],
"language": "python",
"metadata": {},
"outputs": [
{
"html": [
"\n",
"
\n",
" \n",
" \n",
" | \n",
" rank | \n",
" year | \n",
" movie_title | \n",
" imdbid | \n",
" studio | \n",
" total_gross | \n",
" num_theaters_total | \n",
" opening_revenue | \n",
" num_theaters_opening | \n",
" date_open | \n",
" date_close | \n",
" oscar_winner | \n",
" oscar_nom | \n",
"
\n",
" \n",
" \n",
" \n",
" | 0 | \n",
" 1 | \n",
" 2003 | \n",
" The Lord of the Rings: The Return of the King | \n",
" 0167260 | \n",
" NL | \n",
" 377027325 | \n",
" 3703 | \n",
" 72629713 | \n",
" 3703 | \n",
" 17-Dec | \n",
" 3-Jun | \n",
" 1 | \n",
" 1 | \n",
"
\n",
" \n",
" | 1 | \n",
" 2 | \n",
" 2003 | \n",
" Finding Nemo | \n",
" 0266543 | \n",
" BV | \n",
" 339714978 | \n",
" 3425 | \n",
" 70251710 | \n",
" 3374 | \n",
" 30-May | \n",
" 18-Dec | \n",
" 1 | \n",
" 1 | \n",
"
\n",
" \n",
" | 2 | \n",
" 3 | \n",
" 2003 | \n",
" Pirates of the Caribbean: The Curse of the Bla... | \n",
" 0325980 | \n",
" BV | \n",
" 305413918 | \n",
" 3416 | \n",
" 46630690 | \n",
" 3269 | \n",
" 9-Jul | \n",
" 8-Jan | \n",
" 0 | \n",
" 1 | \n",
"
\n",
" \n",
" | 3 | \n",
" 4 | \n",
" 2003 | \n",
" The Matrix Reloaded | \n",
" 0234215 | \n",
" WB | \n",
" 281576461 | \n",
" 3603 | \n",
" 91774413 | \n",
" 3603 | \n",
" 15-May | \n",
" 13-Nov | \n",
" 0 | \n",
" 0 | \n",
"
\n",
" \n",
" | 4 | \n",
" 5 | \n",
" 2003 | \n",
" Bruce Almighty | \n",
" 0315327 | \n",
" Uni. | \n",
" 242829261 | \n",
" 3549 | \n",
" 67953330 | \n",
" 3483 | \n",
" 23-May | \n",
" 11-Dec | \n",
" 0 | \n",
" 0 | \n",
"
\n",
" \n",
"
\n",
"
Movie Studios
"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"def plot_str_hist(df,column_name,xlabel='',ylabel='',title='',xticksize=7):\n",
" total_movies = float(len(df))\n",
" data_pts=[]\n",
" counts=[]\n",
" data=[]\n",
" tickloc=0.5\n",
" barloc=0\n",
" barlocs=[]\n",
" ticklocs=[]\n",
" grouped_data=(df.groupby(column_name)).groups\n",
" for key,val in grouped_data.items():\n",
" data.append((key,len(val)))\n",
" ticklocs.append(tickloc)\n",
" barlocs.append(barloc)\n",
" tickloc+=1\n",
" barloc+=1\n",
" data.sort(key=lambda x:x[1])\n",
" for key,val in data:\n",
" data_pts.append(key)\n",
" counts.append(val/total_movies)\n",
" plt.bar(barlocs, height=counts, width=1,color='r',edgecolor=\"white\")\n",
" plt.xticks(rotation=90,size=xticksize)\n",
" plt.xticks(ticklocs,data_pts)\n",
" plt.xlabel(xlabel)\n",
" plt.ylabel(ylabel)\n",
" plt.title(title)\n",
" histogram_style()\n",
" plt.show()\n",
"\n",
"plot_str_hist(bomj_df,'studio',xlabel='Studio',ylabel='Frequency',title='Studio v. Count for All Movies')\n",
"plot_str_hist(bomj_df[bomj_df.oscar_winner == 1],'studio',xlabel='Studio',ylabel='Frequency',title='Studio v. Count for Winners',xticksize=10)\n",
"plot_str_hist(bomj_df[bomj_df.oscar_nom == 1],'studio',xlabel='Studio',ylabel='Frequency',title='Studio v. Count for Nominees',xticksize=10)"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "display_data",
"png": "iVBORw0KGgoAAAANSUhEUgAAAnAAAAGmCAYAAAD8nLkMAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
"text": [
""
]
},
{
"metadata": {},
"output_type": "display_data",
"png": "iVBORw0KGgoAAAANSUhEUgAAAnAAAAGjCAYAAACsUSi/AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nRU5ErkJggg==\n",
"text": [
""
]
},
{
"metadata": {},
"output_type": "display_data",
"png": "iVBORw0KGgoAAAANSUhEUgAAAnAAAAGjCAYAAACsUSi/AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\noTQwqAFHoVAoFAqF0sD4f9QE9+PdTR0AAAAAAElFTkSuQmCC\n",
"text": [
""
]
}
],
"prompt_number": 170
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Quantitative BOMJ Data
"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"def plot_quant_data(bomj_df,varname,cr):\n",
" \n",
" data = (bomj_df)[varname].values\n",
" cleaned_data = []\n",
" for pt in data:\n",
" try:\n",
" cleaned_data.append(int(pt))\n",
" except:\n",
" pass\n",
" plt.hist(cleaned_data,20,color=cr,normed=True)\n",
" plt.xlabel(varname)\n",
" plt.ylabel('Frequency')\n",
" plt.title(varname + ' v. count for all movies')\n",
" histogram_style()\n",
" plt.show()\n",
" \n",
" win_data = (bomj_df[bomj_df.oscar_winner == 1])[varname].values\n",
" cleaned_win_data = []\n",
" for pt in win_data:\n",
" try:\n",
" cleaned_win_data.append(int(pt))\n",
" except:\n",
" pass \n",
" plt.hist(cleaned_win_data,20,color=cr,normed=True)\n",
" plt.xlabel(varname)\n",
" plt.ylabel('Frequency')\n",
" plt.title(varname + ' v. count for winners')\n",
" histogram_style()\n",
" plt.show()\n",
"\n",
" nom_data = (bomj_df[bomj_df.oscar_nom == 1])[varname].values\n",
" cleaned_nom_data = []\n",
" for pt in nom_data:\n",
" try:\n",
" cleaned_nom_data.append(int(pt))\n",
" except:\n",
" pass \n",
" plt.hist(cleaned_nom_data,20,color=cr,normed=True)\n",
" plt.xlabel(varname)\n",
" plt.ylabel('Frequency')\n",
" plt.title(varname + ' v. count for nominees')\n",
" histogram_style()\n",
" plt.show()\n",
"plot_quant_data(bomj_df,'num_theaters_total','b')\n",
"plot_quant_data(bomj_df,'opening_revenue','b')\n",
"plot_quant_data(bomj_df,'num_theaters_opening','b')"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "display_data",
"png": "iVBORw0KGgoAAAANSUhEUgAAApMAAAGVCAYAAACip3p9AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
"text": [
""
]
},
{
"ename": "AttributeError",
"evalue": "'DataFrame' object has no attribute 'oscar_winner'",
"output_type": "pyerr",
"traceback": [
"\u001b[0;31m---------------------------------------------------------------------------\u001b[0m\n\u001b[0;31mAttributeError\u001b[0m Traceback (most recent call last)",
"\u001b[0;32m\u001b[0m in \u001b[0;36m\u001b[0;34m()\u001b[0m\n\u001b[1;32m 42\u001b[0m \u001b[0mhistogram_style\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 43\u001b[0m \u001b[0mplt\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mshow\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 44\u001b[0;31m \u001b[0mplot_quant_data\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mbomj_df\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m'num_theaters_total'\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m'b'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 45\u001b[0m \u001b[0mplot_quant_data\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mbomj_df\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m'opening_revenue'\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m'b'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 46\u001b[0m \u001b[0mplot_quant_data\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mbomj_df\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m'num_theaters_opening'\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m'b'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
"\u001b[0;32m\u001b[0m in \u001b[0;36mplot_quant_data\u001b[0;34m(bomj_df, varname, cr)\u001b[0m\n\u001b[1;32m 15\u001b[0m \u001b[0mplt\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mshow\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 16\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 17\u001b[0;31m \u001b[0mwin_data\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m(\u001b[0m\u001b[0mbomj_df\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mbomj_df\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0moscar_winner\u001b[0m \u001b[0;34m==\u001b[0m \u001b[0;36m1\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mvarname\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mvalues\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 18\u001b[0m \u001b[0mcleaned_win_data\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 19\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0mpt\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mwin_data\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
"\u001b[0;32m//anaconda/python.app/Contents/lib/python2.7/site-packages/pandas/core/frame.pyc\u001b[0m in \u001b[0;36m__getattr__\u001b[0;34m(self, name)\u001b[0m\n\u001b[1;32m 2086\u001b[0m \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mname\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 2087\u001b[0m raise AttributeError(\"'%s' object has no attribute '%s'\" %\n\u001b[0;32m-> 2088\u001b[0;31m (type(self).__name__, name))\n\u001b[0m\u001b[1;32m 2089\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 2090\u001b[0m \u001b[0;32mdef\u001b[0m \u001b[0m__setattr__\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mname\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mvalue\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
"\u001b[0;31mAttributeError\u001b[0m: 'DataFrame' object has no attribute 'oscar_winner'"
]
}
],
"prompt_number": 168
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"IMDB Reviews & Ratings
"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"In order to understand what the distribution of movies and movie reviewers looks like, we calculated the number of reviews, critics, and movies and then plot the number of reviews per critic."
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"print \"Number of reviews: %i\" % len(imdb_df)\n",
"print \"Number of critics: %i\" % imdb_df.critic.unique().size\n",
"print \"Number of movies: %i\" % imdb_df.title.unique().size"
],
"language": "python",
"metadata": {},
"outputs": [
{
"output_type": "stream",
"stream": "stdout",
"text": [
"Number of reviews: 435277\n",
"Number of critics: 199188"
]
},
{
"output_type": "stream",
"stream": "stdout",
"text": [
"\n",
"Number of movies: 1900\n"
]
}
],
"prompt_number": 6
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"imdb_df.groupby('critic').title.count().hist(log=True, bins=range(20), edgecolor='white')\n",
"plt.xlabel(\"Number of reviews per critic\")\n",
"plt.ylabel(\"N\")\n",
"histogram_style()"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "display_data",
"png": "iVBORw0KGgoAAAANSUhEUgAAAnQAAAGMCAYAAABTQD8mAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nSAAAAABJRU5ErkJggg==\n",
"text": [
""
]
}
],
"prompt_number": 7
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"As one can see, although we have a very large number of reviews and reviews, most of them do not interview that many movies themselves."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Next, we will visualize how the number of reviews changes over the years, and how the average ratings equuivalently change."
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"plt.plot(imdb_df.groupby('year').critic.count().index,imdb_df.groupby('year').critic.count())\n",
"plt.xlabel('Year')\n",
"plt.ylabel('Number of Reviews')\n",
"plt.show()"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "display_data",
"png": "iVBORw0KGgoAAAANSUhEUgAAAo4AAAGJCAYAAAAE47LFAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nI20AAAAASUVORK5CYII=\n",
"text": [
""
]
}
],
"prompt_number": 8
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"From this plot, as expected, there is a sharp increase in the number of reviews beginning in 2003, the start of our data, but strangely enough, there is some (sparse) data for years previous 2002. Let's investigate those."
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"imdb_df[imdb_df['year'] < 2003].groupby('year').title.unique()"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 9,
"text": [
"year\n",
"2001 [Winged Migration, Nowhere in Africa, Monsters...\n",
"2002 [Phone Booth, 28 Days Later, Bend It Like Beck...\n",
"dtype: object"
]
}
],
"prompt_number": 9
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Considering the earliest movie we tried to get was in 2003, it looks like something went wrong. These movies *are* from 2002 and 2001, so these errors are likely from the original year classification. Because we are only interested in years in [2003,2012], we will just remove these entries from the dataframe."
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"# let's remove the entries for years < 2003\n",
"imdb_df = imdb_df[imdb_df['year'] >= 2003]"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 10
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"# Finally, let's try plotting the same graph as above\n",
"plt.plot(imdb_df.groupby('year').critic.count().index,imdb_df.groupby('year').critic.count())\n",
"plt.xlabel('Year')\n",
"plt.xticks(range(2002,2013))\n",
"plt.ylabel('Number of Reviews')\n",
"plt.show()"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "display_data",
"png": "iVBORw0KGgoAAAANSUhEUgAAAo4AAAGJCAYAAAAE47LFAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\niKqEwZGIiIiIqoTBkYiIiIiq5P8BinI1VvXunXkAAAAASUVORK5CYII=\n",
"text": [
""
]
}
],
"prompt_number": 11
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"As expected, we now only have our data from 2003 onwards. Interestingly, the peak of user reviews is in 2005, with the number of reviews exponentially falling off until 2011. Thus, it seems that as IMDB became more widespread and popular, *fewer* people actually have been leaving reviews for movies.\n",
"\n",
"Now that we have an idea of the distribution of the data over the years, let's look at the actual review ratings. On IMDB users have the option of rating movies without actually leaving a review. Here, we compare the overall movie rating to the average rating user critics give as an approximate measure of whether user critics or general users rate movies more harshly."
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"print \"average movie rating: %0.2f / 10\" % (imdb_df.groupby('title')['overall_score'].unique().mean() * 10)\n",
"print \"average user critic rating: %0.2f / 10\" % (imdb_df.groupby('critic')['norm_score'].mean().mean() * 10) "
],
"language": "python",
"metadata": {},
"outputs": [
{
"output_type": "stream",
"stream": "stdout",
"text": [
"average movie rating: 6.27 / 10\n",
"average user critic rating: 6.27 / 10"
]
},
{
"output_type": "stream",
"stream": "stdout",
"text": [
"\n"
]
}
],
"prompt_number": 12
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"From this measure, it appears that the user critics rate the movies *the same* as the average user. Let's try to visualize this a little more explicitly. "
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"#Your code here\n",
"imdb_df.groupby('critic').norm_score.mean().hist(bins=10, edgecolor='w', lw=1)\n",
"plt.xlabel(\"Average rating per critic\")\n",
"plt.ylabel(\"N\")\n",
"histogram_style()"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "display_data",
"png": "iVBORw0KGgoAAAANSUhEUgAAAocAAAGJCAYAAAD49pkPAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAIABJREFUeJzt3Xt0VOW9//HP5AK5TJoIZBJIcgA1yEVBThBJKDeJQQWM\n1lYkEgSloYUTQGRFaa2iIhUNiIBa8QIIAirFtpZLgge0BIZVDJeDQvBChHJJIEiDiSGEyfP7w/L8\nHESUIDOBvF9rzVrNfr6z93fvZ1U/7ts4jDFGAAAAgKQAfzcAAACA+oNwCAAAAItwCAAAAItwCAAA\nAItwCAAAAItwCAAAAItwCAAAAMun4XDSpEkKCAjw+rRo0eI7NXFxcQoLC1OfPn20Y8cOr/Hq6mpl\nZ2crOjpaTqdT6enp2r9/v1fN0aNHlZmZqaioKEVFRWno0KEqLy/3qtm7d68GDhwop9Op6OhojR07\nVjU1NRdmxwEAAC4SPj9z2LZtW5WUlNjP9u3b7djUqVM1ffp0zZ49W5s2bZLL5dKNN96oiooKWzNu\n3DgtW7ZMS5Ys0bp163Ts2DENGDBAtbW1tiYjI0Nbt25VXl6eVq1apc2bNyszM9OOezwe9e/fX5WV\nlSooKNDixYu1dOlSPfDAA745CAAAAPWV8aFHH33UXH311Wccq62tNbGxsWbKlCl2WVVVlYmIiDAv\nvfSSMcaYf//736ZRo0Zm0aJFtuZf//qXCQgIMHl5ecYYY3bs2GEcDofZsGGDrSkoKDAOh8N88skn\nxhhjVqxYYQICAsy+fftszcKFC01ISIj56quvfrodBgAAuMj4/Mzh7t27FRcXp8svv1yDBw9WcXGx\nJKm4uFilpaVKS0uztSEhIerZs6c2bNggSSosLFRNTY1XTXx8vNq1aye32y1JcrvdcjqdSk5OtjUp\nKSkKDw+363G73Wrfvr3i4uJsTVpamqqrq1VYWHjhdh4AAKCe82k47Natm+bPn6+8vDy9/PLLKikp\nUUpKir788kuVlJRIkmJiYry+43K57FhJSYkCAwPVtGlTr5qYmBivmujoaK9xh8PxnfWcvp1mzZop\nMDDQ1gAAADREQb7c2E033WT/99VXX63k5GS1bt1a8+fP1/XXX/+933M4HGddrzHmnHupy3cAAAAu\ndX59lU1YWJg6dOigzz77TM2bN5cklZaWetWUlpYqNjZWkhQbGyuPx6MjR46ctebw4cNe48YYHTp0\nyKvm9O2UlZXJ4/HYGgAAgIbIr+Hw+PHj2rlzp5o3b67WrVsrNjZW+fn5XuMFBQVKSUmRJCUlJSk4\nONirZt++fSoqKrI1ycnJqqiosPcgSt/cY1hZWWlrUlJStHPnTq9X4KxevVqNGzdWUlLSBd1nAACA\n+sxhfHh9dcKECbr11luVkJCgQ4cO6YknnlBBQYG2b9+uhIQEPf3005oyZYrmzp2rxMRETZ48WQUF\nBdq1a5fCw8MlSaNGjdK7776refPmqUmTJho/frzKy8tVWFhoLz/fcsst2rdvn+bMmSNjjLKysnT5\n5Zfrr3/9qySptrZW1157raKjozVt2jSVlZVp2LBhuuOOO/Tcc8/56nAAAADUOz6953D//v0aPHiw\nysrKFB0dreTkZG3cuFEJCQmSpJycHFVVVWn06NE6evSounXrpvz8fBsMJWnGjBkKCgrSoEGDVFVV\npdTUVC1cuNDrvsRFixYpOztb/fr1kySlp6dr9uzZdjwgIEDLly/XqFGj1L17d4WGhmrIkCF65pln\nfHQkAAAA6iefnjkEAABA/cZvKwMAAMAiHAIAAMAiHAIAAMAiHAIAAMAiHAIAAMAiHAIAAMAiHAIA\nAMAiHAIAAMAiHAIAAMAiHAIAAMAiHAIAAMAiHAIAAMAiHAIAAMAiHAIAAMAiHAIAAMAiHAIAAMAi\nHAIAAMAiHAIAAMAiHAIAAEiqqfX4u4V6wWGMMf5uAgAAoD6In/uQv1s4b/uGP3Ve3+fMIQAAACzC\nIQAAACzCIQAAACzCIQAAACzCIQAAACzCIQAAACzCIQAAACzCIQAAACzCIQAAACzCIQAAACzCIQAA\nACzCIQAAACzCIQAAACzCIQAAACzCIQAAACzCIQAAACzCIQAAACzCIQAAACzCIQAAACzCIQAAACzC\nIQAAACzCIQAAACzCIQAAACzCIQAAACzCIQAAACzCIQAAACzCIQAAACzCIQAAACzCIQAAACzCIQAA\nACzCIQAAACzCIQAAACzCIQAAACzCIQAAACzCIQAAACzCIQAAACy/hcM//vGPCggIUHZ2ttfySZMm\nKS4uTmFhYerTp4927NjhNV5dXa3s7GxFR0fL6XQqPT1d+/fv96o5evSoMjMzFRUVpaioKA0dOlTl\n5eVeNXv37tXAgQPldDoVHR2tsWPHqqam5sLsLAAAwEXCL+Fw48aNevnll9WxY0c5HA67fOrUqZo+\nfbpmz56tTZs2yeVy6cYbb1RFRYWtGTdunJYtW6YlS5Zo3bp1OnbsmAYMGKDa2lpbk5GRoa1btyov\nL0+rVq3S5s2blZmZacc9Ho/69++vyspKFRQUaPHixVq6dKkeeOAB3xwAAACAesphjDG+3GB5ebmS\nkpL06quvatKkSbrmmms0c+ZMGWPUokULjRkzRhMnTpQkHT9+XC6XS7m5ucrKylJ5eblcLpfmzZun\nwYMHS5L27dunli1bauXKlUpLS9POnTvVoUMHrV+/XsnJyZKk9evXq0ePHtq1a5cSExO1cuVKDRgw\nQHv37lVcXJwk6Y033tCIESN0+PBhOZ1OXx4SAABQT8TPfcjfLZy3fcOfOq/v+/zMYVZWln71q1+p\nV69e+nYuLS4uVmlpqdLS0uyykJAQ9ezZUxs2bJAkFRYWqqamxqsmPj5e7dq1k9vtliS53W45nU4b\nDCUpJSVF4eHhdj1ut1vt27e3wVCS0tLSVF1drcLCwguz4wAAABeBIF9u7OWXX9bu3bu1aNEiSfK6\npFxSUiJJiomJ8fqOy+XSgQMHbE1gYKCaNm3qVRMTE2O/X1JSoujoaK9xh8Mhl8vlVXP6dpo1a6bA\nwEBbAwAA0BD5LBzu2rVLv//971VQUKDAwEBJkjFGP+aq9rdD5JnU5cq4j6+mAwAAXBR8Fg7dbrfK\nysrUoUMHu8zj8WjdunV66aWX9NFHH0mSSktLFR8fb2tKS0sVGxsrSYqNjZXH49GRI0e8zh6Wlpaq\nV69etubw4cNe2zbG6NChQ17rOXWJ+ZSysjJ5PB5bAwAAGp7zvV/vUuCzcHj77bera9eu9m9jjIYP\nH642bdrod7/7nRITExUbG6v8/HwlJSVJ+uaBlIKCAuXm5kqSkpKSFBwcrPz8fK8HUoqKipSSkiJJ\nSk5OVkVFhdxut73v0O12q7Ky0takpKToySef1P79++19h6tXr1bjxo3ttgEAQMPDAyk+DIeRkZGK\njIz0WhYWFqbLLrtM7du3l/TNa2qmTJmitm3bKjExUZMnT1ZERIQyMjLsOu677z7l5OTI5XKpSZMm\nGj9+vDp16qTU1FRJUrt27XTTTTdp5MiRmjNnjowxGjlypAYOHKjExERJ3zx80qFDBw0dOlTTpk1T\nWVmZcnJylJWVxZPKAACgQfPpAymnczgcXvcT5uTkqKqqSqNHj9bRo0fVrVs35efnKzw83NbMmDFD\nQUFBGjRokKqqqpSamqqFCxd6rWfRokXKzs5Wv379JEnp6emaPXu2HQ8ICNDy5cs1atQode/eXaGh\noRoyZIieeeYZH+w1AABA/eXz9xwCAADUV1xW5reVAQAA8C2EQwAAAFiEQwAAAFiEQwAAAFiEQwAA\nAFiEQwAAAFiEQwAAAFiEQwAAAFiEQwAAAFiEQwAAAFiEQwAAAFiEQwAAAFiEQwAAAFiEQwAAAFiE\nQwAAAFiEQwAAAFiEQwAAAFiEQwAAAFiEQwAAAFiEQwAAAFiEQwAAAFiEQwAAAFiEQwAAAFiEQwAA\nAFiEQwAAAFiEQwAAAFiEQwAAAFiEQwAAAFiEQwAAAFiEQwAAAFiEQwAAAFiEQwAAAFiEQwAAAFiE\nQwAAAFiEQwAAAFiEQwAAAFiEQwAAAFiEQwAAAFiEQwAAAFiEQwAAAFiEQwAAAFiEQwAAAFiEQwAA\nAFiEQwAAAFiEQwAAAFiEQwAAAFiEQwAAAFiEQwAAAFiEQwAAAFiEQwAAAFiEQwAAAFiEQwAAAFiE\nQwAAAFiEQwAAAFiEQwAAAFiEQwAAAFg+DYfPP/+8OnXqpMjISEVGRiolJUUrVqzwqpk0aZLi4uIU\nFhamPn36aMeOHV7j1dXVys7OVnR0tJxOp9LT07V//36vmqNHjyozM1NRUVGKiorS0KFDVV5e7lWz\nd+9eDRw4UE6nU9HR0Ro7dqxqamouzI4DAABcJHwaDhMSEvT0009ry5YtKiws1A033KDbbrtN27Zt\nkyRNnTpV06dP1+zZs7Vp0ya5XC7deOONqqiosOsYN26cli1bpiVLlmjdunU6duyYBgwYoNraWluT\nkZGhrVu3Ki8vT6tWrdLmzZuVmZlpxz0ej/r376/KykoVFBRo8eLFWrp0qR544AHfHQwAAIB6yGGM\nMf5soGnTpnrqqac0YsQItWjRQmPGjNHEiRMlScePH5fL5VJubq6ysrJUXl4ul8ulefPmafDgwZKk\nffv2qWXLllq5cqXS0tK0c+dOdejQQevXr1dycrIkaf369erRo4d27dqlxMRErVy5UgMGDNDevXsV\nFxcnSXrjjTc0YsQIHT58WE6n0z8HAwAA+FX83If83cJ52zf8qfP6vt/uOfR4PFqyZImOHz+unj17\nqri4WKWlpUpLS7M1ISEh6tmzpzZs2CBJKiwsVE1NjVdNfHy82rVrJ7fbLUlyu91yOp02GEpSSkqK\nwsPD7Xrcbrfat29vg6EkpaWlqbq6WoWFhRd0vwEAAOqzIF9vcPv27UpOTlZ1dbVCQ0P11ltv6aqr\nrrLBLSYmxqve5XLpwIEDkqSSkhIFBgaqadOmXjUxMTEqKSmxNdHR0V7jDodDLpfLq+b07TRr1kyB\ngYG2BgAAoCHyeThs27at/u///k/l5eV6++23ddddd2nt2rVn/Y7D4TjreF2ujPv5ajoAAEC95PNw\nGBwcrMsvv1yS1LlzZ23atEnPP/+8HnnkEUlSaWmp4uPjbX1paaliY2MlSbGxsfJ4PDpy5IjX2cPS\n0lL16tXL1hw+fNhrm8YYHTp0yGs9p85UnlJWViaPx2NrAABAw3O+9+tdCnweDk/n8XhUW1ur1q1b\nKzY2Vvn5+UpKSpL0zQMpBQUFys3NlSQlJSUpODhY+fn5Xg+kFBUVKSUlRZKUnJysiooKud1ue9+h\n2+1WZWWlrUlJSdGTTz6p/fv32/sOV69ercaNG9ttAwCAhocHUnwcDh966CENGDBA8fHx+uqrr7Ro\n0SJ98MEHWrVqlaRvXlMzZcoUtW3bVomJiZo8ebIiIiKUkZEhSYqMjNR9992nnJwcuVwuNWnSROPH\nj1enTp2UmpoqSWrXrp1uuukmjRw5UnPmzJExRiNHjtTAgQOVmJgo6ZuHTzp06KChQ4dq2rRpKisr\nU05OjrKyshrck8rVnpOq9lwa73f8WaNQf7cAAMBFz6fhsLS0VEOGDFFJSYkiIyPVqVMnrVq1Sjfe\neKMkKScnR1VVVRo9erSOHj2qbt26KT8/X+Hh4XYdM2bMUFBQkAYNGqSqqiqlpqZq4cKFXvclLlq0\nSNnZ2erXr58kKT09XbNnz7bjAQEBWr58uUaNGqXu3bsrNDRUQ4YM0TPPPOOjI1F/HD9Zoxc/+kBF\nR0v93cp5uSuxi25q2cHfbQAAcNHz+3sO4V/l1VUasWaB3CW7/d3KeXn4ulv0m6t7+rsNAMBFjsvK\n/LYyAAAAvoVwCAAAAItwCAAAAItwCAAAAItwCAAAAItwCAAAAItwCAAAAItwCAAAAItwCAAAAItw\nCAAAAItwCAAAAItwCAAAAItwCAAAAItwCAAAAItwCAAAAItwCAAAAItwCAAAAItwCAAAAItwCAAA\nAItwCAAAAItwCAAAAItwCAAAAItwCAAAAItwCAAAACvobINffvnlj1pJkyZNfpJmAAAA4F9nDYfN\nmjX7wRU4HA55PJ6frCEAAAD4z1nD4Zo1a8643OFwaNWqVZoxY4aCg4MvSGMAAADwvbOGw969e39n\n2ebNm5WTk6N169YpKytLjzzyyIXqDQAAAD72ox9I2b17twYPHqyuXbuqadOm2rFjh2bNmqXo6OgL\n2R8AAAB86AfDYVlZmcaOHat27dqptLRUbrdbb775pq644gpf9AcAAAAfOms4nDx5sq644gq9//77\n+stf/qI1a9bouuuu81VvAAAA8LGz3nP4yCOPKCQkRPHx8XrhhRf04osvyhjjVeNwOPS3v/3tgjYJ\nAAAA3zhrOBw6dKgcDockfScUnnJqHAAAABe/s4bDefPm+agNAAAA1Af8fB4AAAAswiEAAAAswiEA\nAAAswiEAAAAswiEAAAAswiEAAAAswiEAAAAswiEAAAAswiEAAAAswiEAAAAswiEAADgvNbUef7eA\nn9BZf1sZAADghwQHBCp+7kP+buO87Rv+lL9bqBc4cwgAAACLcAgAAACLcAgAAACLcAgAAACLcAgA\ngB/whC/qK55WBgDADy6VJ3wlnvK91HDmEAAAABbhEAAAABbhEAAAABbhEAAAAJZPw+Ef//hHXXfd\ndYqMjJTL5dKtt96qjz/++Dt1kyZNUlxcnMLCwtSnTx/t2LHDa7y6ulrZ2dmKjo6W0+lUenq69u/f\n71Vz9OhRZWZmKioqSlFRURo6dKjKy8u9avbu3auBAwfK6XQqOjpaY8eOVU1NzU+/4wAAABcJn4bD\nDz74QP/zP/8jt9utNWvWKCgoSKmpqTp69KitmTp1qqZPn67Zs2dr06ZNcrlcuvHGG1VRUWFrxo0b\np2XLlmnJkiVat26djh07pgEDBqi2ttbWZGRkaOvWrcrLy9OqVau0efNmZWZm2nGPx6P+/fursrJS\nBQUFWrx4sZYuXaoHHnjANwcDAACgHvLpq2xWrVrl9feCBQsUGRmpDRs2qH///jLGaMaMGZo4caJu\nv/12SdL8+fPlcrm0aNEiZWVlqby8XK+99prmzZunvn372vW0bNlS7733ntLS0rRz507l5eVp/fr1\nuv766yVJL730knr06KFPP/1UiYmJys/P144dO7R3717FxcVJkp5++mmNGDFCU6ZMkdPp9OGRAQAA\nqB/8es/hsWPHVFtbq8suu0ySVFxcrNLSUqWlpdmakJAQ9ezZUxs2bJAkFRYWqqamxqsmPj5e7dq1\nk9vtliS53W45nU4lJyfbmpSUFIWHh9v1uN1utW/f3gZDSUpLS1N1dbUKCwsv3E4DAADUY34Nh2PH\njlXnzp1tiCspKZEkxcTEeNW5XC47VlJSosDAQDVt2tSrJiYmxqsmOjraa9zhcHxnPadvp1mzZgoM\nDLQ1AAAADY3ffiFl/Pjx2rBhgwoKCuRwOH6w/odqjDHn3ENdvgMAAHAp80s4vP/++/XWW29p7dq1\natWqlV0eGxsrSSotLVV8fLxdXlpaasdiY2Pl8Xh05MgRr7OHpaWl6tWrl605fPiw1zaNMTp06JDX\nek5dYj6lrKxMHo/H1jQEkY1D9fbNWf5uAwAapEvpZ+culX25VPbjfPg8HI4dO1Zvv/221q5dqzZt\n2niNtW7dWrGxscrPz1dSUpIk6fjx4yooKFBubq4kKSkpScHBwcrPz9fgwYMlSfv27VNRUZFSUlIk\nScnJyaqoqJDb7baXrN1utyorK21NSkqKnnzySe3fv9/ed7h69Wo1btzYbrshKK+u0og1C+Qu2e3v\nVs7Lw9fdot9c3dPfbQDAObmUflv5UtiXS2k/zodPw+Ho0aO1cOFC/eUvf1FkZKS9ty8iIkLh4eFy\nOBwaN26cpkyZorZt2yoxMVGTJ09WRESEMjIyJEmRkZG67777lJOTI5fLpSZNmmj8+PHq1KmTUlNT\nJUnt2rXTTTfdpJEjR2rOnDkyxmjkyJEaOHCgEhMTJX3z8EmHDh00dOhQTZs2TWVlZcrJyVFWVhZP\nKgMAgAbLp+HwxRdflMPhsK+gOWXSpEl65JFHJEk5OTmqqqrS6NGjdfToUXXr1k35+fkKDw+39TNm\nzFBQUJAGDRqkqqoqpaamauHChV73JS5atEjZ2dnq16+fJCk9PV2zZ8+24wEBAVq+fLlGjRql7t27\nKzQ0VEOGDNEzzzxzIQ8BAABAvebTcPjtl1SfzaOPPqpHH330e8cbNWqkmTNnaubMmd9bExUVpQUL\nFpx1OwkJCXr33Xd/VE8AAAANAb+tDAAAAItwCAAAAItwCAAAAItwCAAAAItwCAAAAItwCAAAAItw\nCAAAAItwCAAAAItwCAAAAItwCAAAAItwCAAAAItwCAAAAItwCAAAAItwCAC4qNTUevzdAnBJC/J3\nAwAAnIvggEDFz33I322ct33Dn/J3C8AZceYQAAAAFuEQAAAAFuEQAAAAFuEQAAAAFuEQAAAAFuEQ\nAAAAFuEQAAAAFuEQAAAAFuEQAAAAFuEQAAAAFuEQAAAAFuEQAAAAFuEQABqImlqPv1sAcBEI8ncD\nAADfCA4IVPzch/zdxnnbN/wpf7cAXNI4cwgAAACLcAgAAACLcAgAAACLcAgAAACLcAgAAACLcAgA\nAACLcAgAAACLcAgAAACLcAgAAACLcAgAAACLcAgAAACLcAgAAACLcAgAAACLcAgAAACLcAgAAACL\ncAgAAACLcAgAAACLcAgAAACLcAgAAACLcAgAAACLcAgAAACLcAgAAACLcAgAP6Cm1uPvFgDAZ4L8\n3QAAbzW1HgUHBPq7jfN2qeyHJAUHBCp+7kP+buO87Rv+lL9bAHARIBwC9QxBBADgT1xWBgAAgEU4\nBHBBcJ8eAFycfHpZ+R//+Idyc3O1efNmHThwQHPnztU999zjVTNp0iS9/PLLOnr0qK6//no9//zz\nat++vR2vrq7WhAkTtGTJElVVValv37564YUXFBcXZ2uOHj2qMWPG6N1335Uk3XrrrZo1a5YiIyNt\nzd69ezV69GitXbtWoaGhysjIUG5uroKDgy/wUQAahkvl8rjEJXIADYtPzxxWVlaqY8eOeu655xQa\nGiqHw+E1PnXqVE2fPl2zZ8/Wpk2b5HK5dOONN6qiosLWjBs3TsuWLdOSJUu0bt06HTt2TAMGDFBt\nba2tycjI0NatW5WXl6dVq1Zp8+bNyszMtOMej0f9+/dXZWWlCgoKtHjxYi1dulQPPPDAhT8IAAAA\n9ZhPzxzefPPNuvnmmyVJw4YN8xozxmjGjBmaOHGibr/9dknS/Pnz5XK5tGjRImVlZam8vFyvvfaa\n5s2bp759+0qSFixYoJYtW+q9995TWlqadu7cqby8PK1fv17XX3+9JOmll15Sjx499OmnnyoxMVH5\n+fnasWOH9u7da884Pv300xoxYoSmTJkip9PpoyMCAABQv9Sbew6Li4tVWlqqtLQ0uywkJEQ9e/bU\nhg0bJEmFhYWqqanxqomPj1e7du3kdrslSW63W06nU8nJybYmJSVF4eHhdj1ut1vt27f3uhSdlpam\n6upqFRYWXtD9xIXDPW4AAJy/evMqm5KSEklSTEyM13KXy6UDBw7YmsDAQDVt2tSrJiYmxn6/pKRE\n0dHRXuMOh0Mul8ur5vTtNGvWTIGBgbYGF59L5R437m8DAPhTvTlzeDan35t4OmPMOa+zLt8BAAC4\n1NWbM4exsbGSpNLSUsXHx9vlpaWldiw2NlYej0dHjhzxOntYWlqqXr162ZrDhw97rdsYo0OHDnmt\n59Ql5lPKysrk8XhsTUMR2ThUb9+c5e82fjKXylk39qP+uVT2hf2oXy6V/ZAunX25VPbjfNSbcNi6\ndWvFxsYqPz9fSUlJkqTjx4+roKBAubm5kqSkpCQFBwcrPz9fgwcPliTt27dPRUVFSklJkSQlJyer\noqJCbrfb3nfodrtVWVlpa1JSUvTkk09q//799r7D1atXq3HjxnbbDUV5dZVGrFkgd8luf7dyXh6+\n7hb95uqel8xlZfajfrlU9oX9qF8ulf2QLp19uZT243z4NBxWVlbq008/lSTV1tZqz5492rp1q5o2\nbaqEhASNGzdOU6ZMUdu2bZWYmKjJkycrIiJCGRkZkqTIyEjdd999ysnJkcvlUpMmTTR+/Hh16tRJ\nqampkqR27drppptu0siRIzVnzhwZYzRy5EgNHDhQiYmJkr55+KRDhw4aOnSopk2bprKyMuXk5Cgr\nK4snlQEAQIPm03C4adMm3XDDDZK+uY/w0Ucf1aOPPqphw4bptddeU05OjqqqqjR69GgdPXpU3bp1\nU35+vsLDw+06ZsyYoaCgIA0aNEhVVVVKTU3VwoULve5LXLRokbKzs9WvXz9JUnp6umbPnm3HAwIC\ntHz5co0aNUrdu3dXaGiohgwZomeeecZHRwIAAKB+8mk47N27t9fLqs/kVGD8Po0aNdLMmTM1c+bM\n762JiorSggULzrqdhIQE+wsqAAAA+MZF8bQyAAAAfINwCAAAAItwCAAAAItwCAAAAItwCAAAAItw\nCAAAAItwCAAAAItwCAAAAItwCAAAAItwCAAAAItwCAAAAItwCAAAAItwCAAAAItwCAAAAItwCAAA\nAItwCAAAAItwCAAAAItwCAAAAItwCAAAAItwCAAAAItwCAAAAItwCAAAAItwCAAAAItwCAAAAItw\nCAAAAItwCAAAAItwCAAAAItwCAAAAItwCAAAAItwCAAAAItwCAAAAItwCAAAAItwCAAAAItwCAAA\nAItwCAAAAItwCAAAAItwCAAAAItwCAAAAItwCAAAAItwCAAAAItwCAAAAItwCAAAAItwCAAAAItw\nCAAAAItwCAAAAItwCAAAAItwCAAAAItwCAAAAItwCAAAAItwCAAAAItwCAAAAItwCAAAAItwCAAA\nAItwCAAAAItwCAAAAItwCAAAAKvBh8MXXnhBrVu3VmhoqLp06aKCggJ/twQAAOA3DTocvvnmmxo3\nbpwefvgXmNzEAAAT8ElEQVRhbd26VSkpKbr55pv1r3/9y9+tAQAA+EWDDofTp0/X8OHDdd999+mq\nq67SzJkz1bx5c7344otede+//75/GsRPorpor79bQB0xdxc35u/ixdxd3M43tzTYcHjixAlt3rxZ\naWlpXsvT0tK0YcMGr2WEw4sb/5C7eDF3Fzfm7+LF3F3cCId1VFZWJo/Ho5iYGK/lLpdLJSUlfuoK\nAADAv4L83QD8K9Dh0OPXD9SxE8f93cp5+a+IJv5uAQCAS4NpoKqrq01QUJBZunSp1/JRo0aZ3r17\ney275557jCQ+fPjw4cOHD596/7nnnnvOKyM5jDFGDVS3bt3UqVMnvfTSS3ZZmzZt9Ktf/UpPPvmk\nHzsDAADwjwZ9WXn8+PHKzMxU165dlZKSoj/96U8qKSnRb37zG3+3BgAA4BcNOhzeeeedOnLkiCZP\nnqyDBw/qmmuu0YoVK5SQkODv1gAAAPyiQV9WBgAAgLcG+yqbbzvXn9Dbvn27evXqpbCwMMXHx+uJ\nJ57wUac4k3OZv/fff1/p6elq0aKFwsPD1alTJ82dO9eH3eLb6vrzlZ9++qkiIiIUERFxgTvE2dRl\n/mbMmKG2bdsqJCRELVq00MSJE33QKU53rnO3YsUKdevWTT/72c8UHR2t2267TZ9++qmPusUp//jH\nP3TrrbcqPj5eAQEBmj9//g9+p06Z5bweZ7kELFmyxAQHB5tXXnnFFBUVmezsbON0Os3evXvPWF9e\nXm5iYmLMoEGDzMcff2yWLl1qIiIizLRp03zcOYw59/mbMmWK+cMf/mA2bNhgiouLzYsvvmiCgoLM\nokWLfNw5znXuTqmurjb//d//bfr3728iIiJ81C1OV5f5u//++02bNm3M3/72N1NcXGy2bt1qVq5c\n6cOuYcy5z92nn35qgoODzYMPPmg+//xzs3XrVtOvXz9z5ZVX+rhzrFixwvz+9783S5cuNWFhYWb+\n/Plnra9rZmnw4bBr164mKyvLa1liYqKZOHHiGetfeOEFExkZaY4fP26XTZ482cTFxV3QPnFm5zp/\nZ3LnnXeaO+6446duDT+grnM3btw4c++995p58+YZp9N5IVvEWZzr/BUVFZng4GBTVFTki/ZwFuc6\nd2+//bYJDAw0tbW1dtmaNWuMw+EwR44cuaC94vs5nc4fDId1zSwN+rLyufyE3ilut1s9evRQ48aN\nveoPHDigPXv2XNB+4a0u83cm5eXlatKEl2j7Ul3nbvny5Vq+fLlmzZolw+3SflOX+fvrX/+qyy+/\nXCtWrNDll1+u1q1ba9iwYTp8+LAvWsZ/1GXuunfvLqfTqZdfflkej0dfffWV5s2bp65du/LPznqu\nrpmlQYfDuvyEXklJyXfqT/3Nz+751k/xE4h///vftWbNGmVlZV2IFvE96jJ3Bw4cUFZWlt544w2F\nhYX5ok18j7rM3+7du7Vnzx699dZbev3117VgwQIVFRVp4MCBBH0fqsvcNW/eXCtWrNDDDz+skJAQ\nRUVF6eOPP9a7777ri5ZxHuqaWRp0OKwLh8Ph7xbwE1m/fr3uvvtuzZo1S126dPF3O/gBmZmZ+u1v\nf6vrrrvO362gDmpra1VdXa0FCxbo5z//uX7+859rwYIF+uc//6kPP/zQ3+3hLHbv3q3bbrtNw4cP\n14cffqj3339fERERuvPOOwn29VxdM0uDDofNmjVTYGCgSktLvZaXlpaqefPmZ/xObGzsd9L2qe/H\nxsZemEZxRnWZv1MKCgp0yy236IknntDIkSMvZJs4g7rM3dq1a/XYY48pODhYwcHBGjFihCorKxUc\nHKxXXnnFF23jP+oyf82bN1dQUJCuvPJKu+zKK69UYGCg9u7de0H7xf9Xl7l76aWXlJCQoKlTp6pT\np07q0aOHFi5cqA8++EBut9sXbaOO6ppZGnQ4bNSokZKSkpSfn++1fPXq1UpJSTnjd5KTk7Vu3TpV\nV1d71cfFxally5YXtF94q8v8Sd+8CuCWW27RY489pjFjxlzoNnEGdZm7jz76SNu2bbOfxx9/XKGh\nodq2bZt++ctf+qJt/Edd5u/nP/+5Tp48qd27d9tlu3fvlsfj4Z+dPlSXuTPGKCDAOy6c+ru2tvbC\nNIqfRJ0zy3k/LnORe/PNN02jRo3MK6+8Ynbs2GHGjBljIiIi7CP9Dz30kOnbt6+tLy8vN7Gxseau\nu+4yH330kfnzn/9sfvazn5np06f7axcatHOdv7Vr15qwsDCTk5NjSkpKzMGDB83BgwfNoUOH/LUL\nDda5zt3p5s6dy9PKfnSu81dbW2uSkpJMr169zJYtW8zmzZtNz549TXJysr92ocE617lbt26dCQgI\nMI8//rj55JNPTGFhoenXr59p2bKl+frrr/21Gw1SRUWF2bJli9myZYsJCwszjz/+uNmyZctPnlka\nfDg05ptHvVu1amUaN25sunTpYtatW2fHhg0bZlq3bu1Vv337dtOzZ08TEhJiWrRoYR5//HFft4xv\nOZf5GzZsmAkICDAOh8Prc/ocwzfO9f973zZ37lzec+hn5zp/Bw8eNL/61a9MRESEcblcZsiQIfyH\nmZ+c69y9/fbbJikpyTidTuNyuUx6errZuXOnr9tu8NauXWv/vfXtf5cNHz7cGPPTZRZ+Pg8AAABW\ng77nEAAAAN4IhwAAALAIhwAAALAIhwAAALAIhwAAALAIhwAAALAIhwAAALAIhwDgY7179+anG/9j\n3rx5ioiI+MG61q1ba/r06T7oCAAvwQZwzjZv3qwuXbooJSVFBQUF/m6n3po3b56ys7P11VdfeS3/\n97//reDgYIWHh/ups/rj+PHjqqioULNmzSRJkyZN0p///Gdt377dq+7IkSMKCwtTaGioP9oEGhTO\nHAI4Z6+88oquu+46bdy4UUVFRRd8eydPnrzg2zgXJ06cOK/vR0VFXdTB8Keaj5MnTyokJMQGw7Np\n2rQpwRDwEcIhgHNSVVWlxYsX67HHHtMNN9ygV1991Y5lZGTol7/8pVd9bW2tEhISNGPGDEmSMUZP\nP/20rrzySoWFhaljx4564403bP0XX3yhgIAALVmyRDfccIPCwsI0Z84cffnllxo8eLASEhIUFham\nq6++WvPmzfPaVmVlpYYOHaqIiAi1aNFCubm5GjBggIYPH25rTpw4oQcffFAJCQkKDw9X165dlZ+f\nf9Z97t27t0aNGqUJEybI5XKpR48ekqTp06erU6dOcjqdio+P169//WuVl5dLkt5//33de++9qqys\nVEBAgAICAvT444/b9WVnZ9v1t2rVSk8++aRGjhypyMhIJSQkKDc316uHTz75RL169VJoaKjat2+v\nVatWyel0av78+d/b97BhwzRw4EBNnjxZsbGxioiI0L333qvjx4971dVlPs7kxIkT+t3vfqdWrVop\nJCREV1xxhWbNmmWPR0BAgFauXKmuXbuqcePGysvL87qsPG/ePD3++OP6+OOP7TF7/fXX7TGaNm2a\n3VZ5ebl++9vfqkWLFvaYvPXWW2eZRQA/2k/9o9AALm2vv/66admypTHGmLffftu4XC5z8uRJY4wx\nK1asMCEhIaa8vNzWr1mzxgQFBZmSkhJjjDG/+93vTNu2bU1eXp754osvzKJFi0x4eLhZvny5McaY\n4uJi43A4TKtWrcyf//xn88UXX5h9+/aZ/fv3m9zcXLNt2zZTXFxs5syZYxo1amT+93//125r5MiR\npmXLlua9994zH3/8sbnrrrtMZGSk/VF6Y4zJyMgwycnJZt26daa4uNjMnj3bNGrUyGzbtu1797lX\nr14mIiLCTJgwwezatcsUFRUZY4yZMWOGWbt2rdmzZ4/54IMPTMeOHU1mZqYxxpgTJ06Y5557zoSH\nh5vS0lJTWlpqKisrjTHG9O7d22RnZ9v1t2zZ0jRt2tQ8//zz5vPPPzezZs0yDofDuN1uY4wxHo/H\ntG/f3qSmpppt27YZt9ttunbtaoKDg838+fO/t+977rnHREREmDvvvNN8/PHHJi8vz8TFxZkxY8bY\nmrrOx5ncddddJj4+3ixbtswUFxebdevWmYULFxpjjFm7dq1xOBymY8eOZvXq1aa4uNgcPnzYzJ07\n1zidTmOMMVVVVWbChAmmbdu29phVVVUZY4xp1aqVmTZtmjHGmNraWpOSkmI6dOhg+87Pzzd/+ctf\nvvdYAPjxCIcAzkmvXr3MY489ZowxpqamxsTExJilS5d6/f3qq6/a+vvuu8/069fPGGNMRUWFCQ0N\nNQUFBV7rHDt2rLnllluMMf8/jEyfPv0He7nrrrvMiBEjjDHGfPXVV6ZRo0bmzTfftOOVlZXmsssu\ns+Hws88+MwEBAWbv3r1e60lPTzejRo066z536tTpB/tZuXKlady4sf3728Hn284UDjMyMrxqEhMT\nzeTJk40xxqxatcoEBQWZAwcO2PENGzYYh8Pxg+Hwsssus6HUGGMWLlxoGjdubL7++uufdD4++eQT\n43A4TF5e3hnHT4XDZcuWeS0//Rg9+uij5uqrr/7O978dDvPz801AQIAN6QB+WkH+PnMJ4OLx2Wef\naf369VqwYIEkKSgoSPfcc49effVV3XHHHQoKCtKgQYP0xhtv6N5771V1dbWWLVummTNnSpJ27Nih\n48ePq1+/fnI4HHa9NTU1at26tde2unTp4vW3x+PRU089pTfffFMHDhxQdXW1Tpw4oT59+kiSPv/8\nc9XU1Khr1672O6cuP5+yefNmGWPUvn17r3VXV1erb9++37vfDodDSUlJ31m+Zs0a/fGPf1RRUZHK\ny8vl8XhUU1OjkpISxcbGnvVYnr7+jh07ei1r0aKFDh8+LEkqKipSixYt1Lx5czvepUsXBQT88J1B\nHTt2VFhYmP27W7duOnHihD7//HNVVVXVeT5Ot2XLFgUEBNj5+D4/tJ4fY8uWLWrevLmuuuqq814X\ngO8iHAL40V555RV5PB5dfvnldpn5zwsP9u3bp/j4eA0ZMkTJyck6cOCANm7cqBMnTugXv/iFpG/u\nP5Skv//97/qv//ovr3UHBwd7/X36Axu5ubmaPn26Zs6cqWuuuUZOp1MTJ060Aer7mG+9kKG2tlYO\nh0Mffvjhd7b3Qw87nN7Pnj171L9/f40cOVKTJ09W06ZNVVhYqMGDB9fpgZXT+3E4HPZ4nQ9zlhdS\nnM981NXF/CAO0FAQDgH8KCdPntT8+fP11FNPacCAAXa5MUaZmZmaO3eu/vCHP+i6667TlVdeqcWL\nF2vDhg267bbb7Jmr9u3bq3Hjxvriiy/Uu3fvc9p+QUGBbr31Vt199912u7t27VKTJk0kSVdccYWC\ng4P1z3/+U61atZIkff311/roo4+UmJgoSercubOMMTp48OA5b/90H374oWpqavTss8/as25/+9vf\nvGoaNWokj8dzXtuRpLZt2+rAgQM6ePCgPXv44Ycf/qjwuH37dn399dd2DjZu3KhGjRrpiiuu0MmT\nJ+s8H6e79tprVVtbqzVr1qhfv351Xs+POWadO3fWwYMHVVRUpLZt29Z5WwDOjHAI4EdZvny5jhw5\nol//+te67LLLvMbuuusu/elPf9If/vAHSdLdd9+tl19+WXv27NE777xj6yIiIjRhwgRNmDBBxhj1\n6NFDFRUV2rhxowIDA/XrX//6e7d/1VVX6c0339T69evVtGlTzZo1S1988YXtxel06t5779WDDz6o\nZs2aKTY2VpMnT5Yxxoa3Nm3a6O6779awYcM0bdo0de7cWV9++aXef/99XXHFFbr99tvPuG3zzf3Z\nXsvatGmj2tpaPfvss7r99tu1ceNGPffcc141rVq10vHjx/Xee+/p2muvVXh4uEJDQ8+4vrNtMy0t\nTVdddZXuuece5ebm6uuvv9b48eMVFBTkdTn4TE6ePKl7771XjzzyiPbv36+HHnpIWVlZ9kxpXefj\ndG3atNGdd96pESNG6LnnnlPnzp21b98+7dmzR0OGDPnR62ndurX27NmjLVu2KCEhQT/72c/UqFEj\nr5rU1FRdf/31uuOOO/Tss88qMTFRn332mb7++mulp6f/6G0BODNeZQPgR3nttdd0ww03fCcYStIv\nf/lL7dmzR++9954kaciQIfrkk08UFRWltLQ0r9onnnhCkyZNUm5urq6++mqlpaXpnXfe8bpUfabA\n8/DDD6tr1666+eab1atXL0VEROjuu+/2qs3NzVWPHj106623qm/fvurUqZO6dOmikJAQWzN37lwN\nHz5cOTk5ateunQYOHKiCggJ7tvFMHA7Hd3q65ppr9Nxzz2n69Onq0KGDXnvtNeXm5nrVpaSk6De/\n+Y0GDx4sl8ulZ5555nvXd7ZtOhwOvfPOO6qurlbXrl01fPhw/f73v5fD4fDatzOto1evXurQoYP6\n9OmjX/ziF0pNTdXTTz9ta+o6H2fy+uuvKyMjQ2PGjFG7du00fPhwHTt27AfX8+3ld9xxh2655Rb1\n7dtXLpdLS5YsOWP9ypUr1b17dw0ZMkTt27fX/fffr5qamh/VJ4Cz4xdSAFyyqqur1bJlSz344IO6\n//77/d3OT2rbtm3q3LmzCgsL1blz5zPWDBs2TEeOHNG7777r4+4AXMy4rAzgkrF161bt2LFDXbt2\n1VdffaWpU6eqsrJSgwYN8ndr5+2dd95ReHi4EhMT9cUXX2j8+PG69tprvzcYAkBdEQ4BXFKeffZZ\n7dq1S0FBQercubP+8Y9/qEWLFv5u67xVVFTooYce0r/+9S9ddtll6tOnj5599tmzfufHXL4GgNNx\nWRkAAAAWD6QAAADAIhwCAADAIhwCAADAIhwCAADAIhwCAADAIhwCAADA+n8ybQhkAam39wAAAABJ\nRU5ErkJggg==\n",
"text": [
""
]
}
],
"prompt_number": 13
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"From this histogram, it appears we have an almost bi-modal distribution where the highest frequency averages are at the polar ends. Considering that a user would be more likely to write a review for a movie if s/he absolutely hated or loved it, this isn't that surprising because then they too would rate it at the extremes. \n",
"
"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Lastly, before we begin out statistical analysis, we must only include reviews that are written in the time *before* the movie was considered for an oscar, else our data would be unnecesarily skewed because after an award does (or does not) win an award, critics may be more inclined to give it a certain review type. In this way, we try to eliminate this particular type of bias here."
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"# first let's convert the date column to an actual date object\n",
"imdb_df['date'] = pd.to_datetime(imdb_df['date'])\n",
"def less_than_date(row):\n",
" return row['date'] < pd.to_datetime('January 1 ' + str(row['year'] + 1))\n",
"imdb_df = imdb_df[imdb_df.apply(less_than_date,axis=1)]"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 6
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"print \"Number of reviews: %i\" % len(imdb_df)\n",
"print \"Number of critics: %i\" % imdb_df.critic.unique().size\n",
"print \"Number of movies: %i\" % imdb_df.title.unique().size"
],
"language": "python",
"metadata": {},
"outputs": [
{
"output_type": "stream",
"stream": "stdout",
"text": [
"Number of reviews: 242533\n",
"Number of critics: 130504"
]
},
{
"output_type": "stream",
"stream": "stdout",
"text": [
"\n",
"Number of movies: 1852\n"
]
}
],
"prompt_number": 10
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Rotten Tomatoes Reviews & Ratings
"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Now we repeat the process for the reviews scraped from Rotten Tomatoes. In order to understand what the distribution of movies and movie reviewers looks like, we calculated the number of reviews, critics, and movies and then plot the number of reviews per critic."
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"print \"Number of reviews: %i\" % len(rotten_df)\n",
"print \"Number of critics: %i\" % rotten_df.critic.unique().size\n",
"print \"Number of movies: %i\" % rotten_df.title.unique().size"
],
"language": "python",
"metadata": {},
"outputs": [
{
"output_type": "stream",
"stream": "stdout",
"text": [
"Number of reviews: 60306\n",
"Number of critics: 876\n",
"Number of movies: 1812\n"
]
}
],
"prompt_number": 73
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"rotten_df.groupby('critic').title.count().hist(log=True, bins=range(20), edgecolor='white')\n",
"plt.xlabel(\"Number of reviews per critic\")\n",
"plt.ylabel(\"N\")\n",
"histogram_style()"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "display_data",
"png": "iVBORw0KGgoAAAANSUhEUgAAAnQAAAGMCAYAAABTQD8mAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nQmCC\n",
"text": [
""
]
}
],
"prompt_number": 74
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"The histogram demonstrates that most critics write between 0 and 5 reviews, and that in general critics do not write more than 20 reviews. In general there is a negative trend: more reviewers are likely to write few reviews, and less reviewers are likely to write many reviews.\n",
"\n",
"As we did before, next we will visualize how the number of reviews changes over the years, and how the average ratings equivalently change."
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"plt.plot(rotten_df.groupby('year').critic.count().index,rotten_df.groupby('year').critic.count())\n",
"plt.xlabel('Year')\n",
"plt.ylabel('Number of Reviews')\n",
"plt.show()"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "display_data",
"png": "iVBORw0KGgoAAAANSUhEUgAAAoYAAAGJCAYAAAAXNPIxAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAAAASUVORK5CYII=\n",
"text": [
""
]
}
],
"prompt_number": 76
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Movies that came out in 2003 and 2006 appear to have received the most reviews, whereas movies that came out in 2009 and 2010 received the least reviews. In general there are never less than 5000 and never more than 7000 reviews for all movies of a specific year. \n",
"\n",
"Now let's look at the actual review ratings. Here, we compare the overall movie rating to the average rating user critics give as an approximate measure of whether user critics or general users rate movies more harshly. Note that on Rotten Tomatoes critics can rate movies without actually leaving a review. Later, when analyzing the reviews themselves, we delete the rows that don't have reviews from our data frame."
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"print \"average movie rating: %0.2f / 10\" % (rotten_df.groupby('title')['overall_score'].unique().mean() * 10)\n",
"print \"average user critic rating: %0.2f / 10\" % (rotten_df.groupby('critic')['norm_score'].mean().mean() * 10) "
],
"language": "python",
"metadata": {},
"outputs": [
{
"output_type": "stream",
"stream": "stdout",
"text": [
"average movie rating: 5.16 / 10\n",
"average user critic rating: 5.06 / 10\n"
]
}
],
"prompt_number": 77
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"It seems that the average rating a movie receives is pretty close to the average score that the critics in the dataframe give. This implies that the score a movie receives by critics is likely to be similar to the score it will receive overall, which means the critics ratings that are included in the dataframe are a good measure for the overall score from this website. \n",
"\n",
"Now we will look at the average rating that critics give in a histogram"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"#Your code here\n",
"rotten_df.groupby('critic').norm_score.mean().hist(bins=10, edgecolor='w', lw=1)\n",
"plt.xlabel(\"Average rating per critic\")\n",
"plt.ylabel(\"N\")\n",
"histogram_style()"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "display_data",
"png": "iVBORw0KGgoAAAANSUhEUgAAAnUAAAGJCAYAAADsT8WrAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nACxA1AEAAFiAqAMAALDA/wEMmS6aRNvugAAAAABJRU5ErkJggg==\n",
"text": [
""
]
}
],
"prompt_number": 79
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"From this histogram, we see that the highest frequency averages are at the polar ends, as it was for the IMDB data set: most critics gave either a 0 or a 1. The third most frequent score is right in the middle at 0.5. Users are indeed likely to write a review if they love or hate the movie. On the other hand, it seems that if they neither hate it or love it, they will give it a 50% rather than thinking too hard if it's on the better or worse end. \n",
"\n",
"\n",
"Next we clear out reviews for a movie that occurred after the oscar awards were given, just as we did with the IMDB reviews. "
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"#def less_than_date(row):\n",
"# return row['date'] < pd.to_datetime('January 1 ' + str(row['year'] + 1))\n",
"\n",
"rotten_df['date'] = pd.to_datetime(rotten_df['date'])\n",
"rotten_df = rotten_df[rotten_df.apply(less_than_date,axis=1)]\n",
"rotten_df.head()"
],
"language": "python",
"metadata": {},
"outputs": [
{
"html": [
"\n",
"
\n",
" \n",
" \n",
" | \n",
" critic | \n",
" date | \n",
" id | \n",
" norm_score | \n",
" overall_score | \n",
" quote | \n",
" source | \n",
" title | \n",
" year | \n",
"
\n",
" \n",
" \n",
" \n",
" | 6 | \n",
" Bill Muller | \n",
" 2003-12-18 00:00:00 | \n",
" 167260 | \n",
" 1 | \n",
" 0.94 | \n",
" Not only has Jackson boldly and faithfully bro... | \n",
" Arizona Republic | \n",
" The Lord of the Rings: The Return of the King | \n",
" 2003 | \n",
"
\n",
" \n",
" | 7 | \n",
" Scott Foundas | \n",
" 2003-12-18 00:00:00 | \n",
" 167260 | \n",
" 1 | \n",
" 0.94 | \n",
" Both the longest film in the series and the mo... | \n",
" L.A. Weekly | \n",
" The Lord of the Rings: The Return of the King | \n",
" 2003 | \n",
"
\n",
" \n",
" | 8 | \n",
" Peter Howell | \n",
" 2003-12-17 00:00:00 | \n",
" 167260 | \n",
" 1 | \n",
" 0.94 | \n",
" Completes the picture, magnificently so, and a... | \n",
" Toronto Star | \n",
" The Lord of the Rings: The Return of the King | \n",
" 2003 | \n",
"
\n",
" \n",
" | 9 | \n",
" Roger Moore | \n",
" 2003-12-17 00:00:00 | \n",
" 167260 | \n",
" 1 | \n",
" 0.94 | \n",
" The thrilling conclusion to what has become th... | \n",
" Orlando Sentinel | \n",
" The Lord of the Rings: The Return of the King | \n",
" 2003 | \n",
"
\n",
" \n",
" | 10 | \n",
" Liam Lacey | \n",
" 2003-12-17 00:00:00 | \n",
" 167260 | \n",
" 1 | \n",
" 0.94 | \n",
" A movie for our time, perhaps for all time, mo... | \n",
" Globe and Mail | \n",
" The Lord of the Rings: The Return of the King | \n",
" 2003 | \n",
"
\n",
" \n",
"
\n",
"
Statistical Analysis
\n",
"\n",
"We will now use a Naive Bayes classifier to build a prediction model for whether a movie will be nominated for an oscar. From there we will try to pick from among the nominees to get a winner."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"As we can see, this filter reduced the number of reviews almost by half, but we also lost some movies."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Before we begin our analysis, we should first add to our dataframe information about the results of the oscar nominations and winners. Here we add two columns to our dataframe: one for the winner, and one for the nomination. For preliminary analysis, we treat all movie-speicifc oscars as the same."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"IMDB Statistical Analysis
"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Because we store the imdb ids as strings in the oscars dataframe, we need to make sure we have a \n",
"consistent representation in the imdb review dataframe. Because we read in from a csv, pandas\n",
"automatically turns the ids into ints, so we have to do the editting here."
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"imdb_df['id'] = imdb_df['id'].map(lambda id: str(id).zfill(7))"
],
"language": "python",
"metadata": {},
"outputs": [
{
"ename": "NameError",
"evalue": "name 'imdb_df' is not defined",
"output_type": "pyerr",
"traceback": [
"\u001b[0;31m---------------------------------------------------------------------------\u001b[0m\n\u001b[0;31mNameError\u001b[0m Traceback (most recent call last)",
"\u001b[0;32m\u001b[0m in \u001b[0;36m\u001b[0;34m()\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0mimdb_df\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'id'\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mimdb_df\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'id'\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmap\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;32mlambda\u001b[0m \u001b[0mid\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mstr\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mid\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mzfill\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;36m7\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
"\u001b[0;31mNameError\u001b[0m: name 'imdb_df' is not defined"
]
}
],
"prompt_number": 171
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"add_oscar_data(imdb_df)\n",
"imdb_df.head()"
],
"language": "python",
"metadata": {},
"outputs": [
{
"html": [
"\n",
"
\n",
" \n",
" \n",
" | \n",
" critic | \n",
" date | \n",
" id | \n",
" norm_score | \n",
" overall_score | \n",
" quote | \n",
" source | \n",
" title | \n",
" year | \n",
" oscar_winner | \n",
" oscar_nom | \n",
"
\n",
" \n",
" \n",
" \n",
" | 0 | \n",
" GuyCC | \n",
" 2003-12-17 00:00:00 | \n",
" 0167260 | \n",
" 1.0 | \n",
" 0.89 | \n",
" Peter Jackson has done it. He has created an ... | \n",
" IMDB | \n",
" The Lord of the Rings: The Return of the King | \n",
" 2003 | \n",
" 1 | \n",
" 1 | \n",
"
\n",
" \n",
" | 1 | \n",
" (muffinking66@hotmail.com) | \n",
" 2003-12-16 00:00:00 | \n",
" 0167260 | \n",
" 1.0 | \n",
" 0.89 | \n",
" Saying that this film starts where `Two Towers... | \n",
" IMDB | \n",
" The Lord of the Rings: The Return of the King | \n",
" 2003 | \n",
" 1 | \n",
" 1 | \n",
"
\n",
" \n",
" | 3 | \n",
" eiratan | \n",
" 2003-12-17 00:00:00 | \n",
" 0167260 | \n",
" 1.0 | \n",
" 0.89 | \n",
" As a movie watcher, I tend to become bored wit... | \n",
" IMDB | \n",
" The Lord of the Rings: The Return of the King | \n",
" 2003 | \n",
" 1 | \n",
" 1 | \n",
"
\n",
" \n",
" | 5 | \n",
" Chris_Docker | \n",
" 2003-12-17 00:00:00 | \n",
" 0167260 | \n",
" 0.9 | \n",
" 0.89 | \n",
" Feeling weary and battle-worn, I have just sta... | \n",
" IMDB | \n",
" The Lord of the Rings: The Return of the King | \n",
" 2003 | \n",
" 1 | \n",
" 1 | \n",
"
\n",
" \n",
" | 6 | \n",
" eviltimmy | \n",
" 2003-12-20 00:00:00 | \n",
" 0167260 | \n",
" 1.0 | \n",
" 0.89 | \n",
" This is the final movie in the Lord of the Rin... | \n",
" IMDB | \n",
" The Lord of the Rings: The Return of the King | \n",
" 2003 | \n",
" 1 | \n",
" 1 | \n",
"
\n",
" \n",
"
\n",
"
Rotten Tomatoes Statistical Analysis
"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Now we repeat the process for the Rotten Tomatoes reviews, first adding columns for oscar winner or oscar nominee. \n"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"rotten_df['id'] = rotten_df['id'].map(lambda id: str(id).zfill(7))\n",
"add_oscar_data(rotten_df)\n",
"rotten_df.head()"
],
"language": "python",
"metadata": {},
"outputs": [
{
"html": [
"\n",
"
\n",
" \n",
" \n",
" | \n",
" critic | \n",
" date | \n",
" id | \n",
" norm_score | \n",
" overall_score | \n",
" quote | \n",
" source | \n",
" title | \n",
" year | \n",
" oscar_winner | \n",
" oscar_nom | \n",
"
\n",
" \n",
" \n",
" \n",
" | 62908 | \n",
" Joe Neumaier | \n",
" 2012-11-08 00:00:00 | \n",
" 1579361 | \n",
" 1 | \n",
" 0.96 | \n",
" You won't want to look away - not only because... | \n",
" New York Daily News | \n",
" Chasing Ice | \n",
" 2012 | \n",
" 0 | \n",
" 1 | \n",
"
\n",
" \n",
" | 62909 | \n",
" Linda Barnard | \n",
" 2012-11-08 00:00:00 | \n",
" 1579361 | \n",
" 1 | \n",
" 0.96 | \n",
" Considering the gravity of the situation there... | \n",
" Toronto Star | \n",
" Chasing Ice | \n",
" 2012 | \n",
" 0 | \n",
" 1 | \n",
"
\n",
" \n",
" | 62910 | \n",
" Chuck Wilson | \n",
" 2012-11-07 00:00:00 | \n",
" 1579361 | \n",
" 1 | \n",
" 0.96 | \n",
" Jeff Orlowski's beautiful yet sobering documen... | \n",
" Village Voice | \n",
" Chasing Ice | \n",
" 2012 | \n",
" 0 | \n",
" 1 | \n",
"
\n",
" \n",
" | 62911 | \n",
" Justin Chang | \n",
" 2012-01-30 00:00:00 | \n",
" 1579361 | \n",
" 1 | \n",
" 0.96 | \n",
" The case for climate change is mounted in visu... | \n",
" Variety | \n",
" Chasing Ice | \n",
" 2012 | \n",
" 0 | \n",
" 1 | \n",
"
\n",
" \n",
" | 62912 | \n",
" John DeFore | \n",
" 2012-01-26 00:00:00 | \n",
" 1579361 | \n",
" 1 | \n",
" 0.96 | \n",
" Doc about photographing glaciers makes a compe... | \n",
" Hollywood Reporter | \n",
" Chasing Ice | \n",
" 2012 | \n",
" 0 | \n",
" 1 | \n",
"
\n",
" \n",
"
\n",
"
Isolating a Training and Validation Set: IMDB
"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"\"\"\"\n",
"Function\n",
"--------\n",
"split_test_train_years:\n",
" randomly splits a movie dataframe by years into training and testing groups\n",
"\n",
"Parameters\n",
"----------\n",
"df:\n",
" Dataframe to be split\n",
"yr_lst:\n",
" a list of the years on which to split\n",
"\n",
"Returns\n",
"-------\n",
"test_yrs, train_years:\n",
" A tuple of the split dataframe\n",
"\"\"\"\n",
"\n",
"def split_test_train_years(df,yr_lst):\n",
" test_yrs = random.sample(yr_lst, len(yr_lst)/2)\n",
" train_yrs = list(set(yr_lst) - set(test_yrs))\n",
" print 'Test years: ' + str(test_yrs)\n",
" print 'Train years: ' + str(train_yrs)\n",
" test_mask = [x in test_yrs for x in df.year]\n",
" train_mask = [x in train_yrs for x in df.year]\n",
" return df[test_mask], df[train_mask]"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 172
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"bomj_testdf,bomj_traindf = split_test_train_years(bomj_df,range(2003,2013))"
],
"language": "python",
"metadata": {},
"outputs": [
{
"output_type": "stream",
"stream": "stdout",
"text": [
"Test years: [2011, 2012, 2010, 2003, 2007]\n",
"Train years: [2008, 2009, 2004, 2005, 2006]\n"
]
}
],
"prompt_number": 173
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"bomj_testdf.head()"
],
"language": "python",
"metadata": {},
"outputs": [
{
"html": [
"\n",
"
\n",
" \n",
" \n",
" | \n",
" rank | \n",
" year | \n",
" movie_title | \n",
" imdbid | \n",
" studio | \n",
" total_gross | \n",
" num_theaters_total | \n",
" opening_revenue | \n",
" num_theaters_opening | \n",
" date_open | \n",
" date_close | \n",
" oscar_winner | \n",
" oscar_nom | \n",
"
\n",
" \n",
" \n",
" \n",
" | 0 | \n",
" 1 | \n",
" 2003 | \n",
" The Lord of the Rings: The Return of the King | \n",
" 0167260 | \n",
" NL | \n",
" 377027325 | \n",
" 3703 | \n",
" 72629713 | \n",
" 3703 | \n",
" 17-Dec | \n",
" 3-Jun | \n",
" 1 | \n",
" 1 | \n",
"
\n",
" \n",
" | 1 | \n",
" 2 | \n",
" 2003 | \n",
" Finding Nemo | \n",
" 0266543 | \n",
" BV | \n",
" 339714978 | \n",
" 3425 | \n",
" 70251710 | \n",
" 3374 | \n",
" 30-May | \n",
" 18-Dec | \n",
" 1 | \n",
" 1 | \n",
"
\n",
" \n",
" | 2 | \n",
" 3 | \n",
" 2003 | \n",
" Pirates of the Caribbean: The Curse of the Bla... | \n",
" 0325980 | \n",
" BV | \n",
" 305413918 | \n",
" 3416 | \n",
" 46630690 | \n",
" 3269 | \n",
" 9-Jul | \n",
" 8-Jan | \n",
" 0 | \n",
" 1 | \n",
"
\n",
" \n",
" | 3 | \n",
" 4 | \n",
" 2003 | \n",
" The Matrix Reloaded | \n",
" 0234215 | \n",
" WB | \n",
" 281576461 | \n",
" 3603 | \n",
" 91774413 | \n",
" 3603 | \n",
" 15-May | \n",
" 13-Nov | \n",
" 0 | \n",
" 0 | \n",
"
\n",
" \n",
" | 4 | \n",
" 5 | \n",
" 2003 | \n",
" Bruce Almighty | \n",
" 0315327 | \n",
" Uni. | \n",
" 242829261 | \n",
" 3549 | \n",
" 67953330 | \n",
" 3483 | \n",
" 23-May | \n",
" 11-Dec | \n",
" 0 | \n",
" 0 | \n",
"
\n",
" \n",
"
\n",
"
![](\"http://i.imgur.com/ea5B6zr.png\")
\n",
"\n",
"In words, we want to:\n",
"\n",
"* Take a collection of examples, and compute the freshness probability for each using `clf.predict_proba`\n",
"* Gather examples into bins of similar freshness probability (the diagram shows 5 groups -- you should use something closer to 20)\n",
"* For each bin, count the number of examples in that bin, and compute the fraction of examples in the bin which are fresh\n",
"* In the upper plot, graph the expected P(Fresh) (x axis) and observed freshness fraction (Y axis). Estimate the uncertainty in observed freshness fraction $F$ via the [equation](http://en.wikipedia.org/wiki/Binomial_proportion_confidence_interval) $\\sigma = \\sqrt{F (1-F) / N}$\n",
"* Overplot the line y=x. This is the trend we would expect if the model is calibrated\n",
"* In the lower plot, show the number of examples in each bin\n",
"\n",
"**Hints**\n",
"\n",
"The output of `clf.predict_proba(X)` is a `(N example, 2)` array. The first column gives the probability $P(Y=0)$ or $P(Rotten)$, and the second gives $P(Y=1)$ or $P(Fresh)$.\n",
"\n",
"The above image is just a guideline -- feel free to explore other options!"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"\"\"\"\n",
"Function\n",
"--------\n",
"calibration_plot\n",
"\n",
"Builds a plot like the one above, from a classifier and review data\n",
"\n",
"Inputs\n",
"-------\n",
"clf : Classifier object\n",
" A MultinomialNB classifier\n",
"X : (Nexample, Nfeature) array\n",
" The bag-of-words data\n",
"Y : (Nexample) integer array\n",
" 1 if a review is Fresh\n",
"\"\"\" \n",
"#your code here\n",
"\n",
"def calibration_plot(clf, xtest, ytest):\n",
" prob = clf.predict_proba(xtest)[:, 1]\n",
" outcome = ytest\n",
" data = pd.DataFrame(dict(prob=prob, outcome=outcome))\n",
"\n",
" #group outcomes into bins of similar probability\n",
" bins = np.linspace(0, 1, 20)\n",
" cuts = pd.cut(prob, bins)\n",
" binwidth = bins[1] - bins[0]\n",
" \n",
" #freshness ratio and number of examples in each bin\n",
" cal = data.groupby(cuts).outcome.agg(['mean', 'count'])\n",
" cal['pmid'] = (bins[:-1] + bins[1:]) / 2\n",
" cal['sig'] = np.sqrt(cal.pmid * (1 - cal.pmid) / cal['count'])\n",
" \n",
" #the calibration plot\n",
" ax = plt.subplot2grid((3, 1), (0, 0), rowspan=2)\n",
" p = plt.errorbar(cal.pmid, cal['mean'], cal['sig'])\n",
" plt.plot(cal.pmid, cal.pmid, linestyle='--', lw=1, color='k')\n",
" plt.ylabel(\"Empirical P(Fresh)\")\n",
" remove_border(ax)\n",
" \n",
" #the distribution of P(fresh)\n",
" ax = plt.subplot2grid((3, 1), (2, 0), sharex=ax)\n",
" \n",
" plt.bar(left=cal.pmid - binwidth / 2, height=cal['count'],\n",
" width=.95 * (bins[1] - bins[0]),\n",
" fc=p[0].get_color())\n",
" \n",
" plt.xlabel(\"Predicted P(Fresh)\")\n",
" remove_border()\n",
" plt.ylabel(\"Number\")"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 17
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"calibration_plot(clf, xtest, ytest)"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "display_data",
"png": "iVBORw0KGgoAAAANSUhEUgAAAn4AAAGJCAYAAAAUrz5cAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nrkJggg==\n",
"text": [
""
]
}
],
"prompt_number": 18
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"**3.5** We might say a model is *over-confident* if the freshness fraction is usually closer to 0.5 than expected (that is, there is more uncertainty than the model predicted). Likewise, a model is *under-confident* if the probabilities are usually further away from 0.5. Is this model generally over- or under-confident? "
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"**Answer** *This model is over-confident. For a properly calibrated model, we would expect ~10% of the P(Fresh)~0.1 reviews to actually be fresh. However, the actual freshness rate is closer to 30%. Likewise, for reviews where P(Fresh) ~0.9, the actuall freshness fraction is closer to 0.7. In other words, there is more uncertainty in the outcome than implied by the model.*"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"#### Cross Validation\n",
"\n",
"Our classifier has a few free parameters. The two most important are:\n",
"\n",
" 1. The `min_df` keyword in `CountVectorizer`, which will ignore words which appear in fewer than `min_df` fraction of reviews. Words that appear only once or twice can lead to overfitting, since words which occur only a few times might correlate very well with Fresh/Rotten reviews by chance in the training dataset.\n",
" \n",
" 2. The [`alpha` keyword](http://scikit-learn.org/stable/modules/naive_bayes.html#multinomial-naive-bayes) in the Bayesian classifier is a \"smoothing parameter\" -- increasing the value decreases the sensitivity to any single feature, and tends to pull prediction probabilities closer to 50%. \n",
"\n",
"As discussed in lecture and HW2, a common technique for choosing appropriate values for these parameters is **cross-validation**. Let's choose good parameters by maximizing the cross-validated log-likelihood.\n",
"\n",
"**3.6** Using `clf.predict_logproba`, write a function that computes the log-likelihood of a dataset"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"\"\"\"\n",
"Function\n",
"--------\n",
"log_likelihood\n",
"\n",
"Compute the log likelihood of a dataset according to \n",
"a bayesian classifier. \n",
"The Log Likelihood is defined by\n",
"\n",
"L = Sum_fresh(logP(fresh)) + Sum_rotten(logP(rotten))\n",
"\n",
"Where Sum_fresh indicates a sum over all fresh reviews, \n",
"and Sum_rotten indicates a sum over rotten reviews\n",
" \n",
"Parameters\n",
"----------\n",
"clf : Bayesian classifier\n",
"x : (nexample, nfeature) array\n",
" The input data\n",
"y : (nexample) integer array\n",
" Whether each review is Fresh\n",
"\"\"\"\n",
"#your code here\n",
"\n",
"def log_likelihood(clf, x, y):\n",
" prob = clf.predict_log_proba(x)\n",
" rotten = y == 0\n",
" fresh = ~rotten\n",
" return prob[rotten, 0].sum() + prob[fresh, 1].sum()"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 19
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Here's a function to estimate the cross-validated value of a scoring function, given a classifier and data"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"from sklearn.cross_validation import KFold\n",
"\n",
"def cv_score(clf, x, y, score_func):\n",
" \"\"\"\n",
" Uses 5-fold cross validation to estimate a score of a classifier\n",
" \n",
" Inputs\n",
" ------\n",
" clf : Classifier object\n",
" x : Input feature vector\n",
" y : Input class labels\n",
" score_func : Function like log_likelihood, that takes (clf, x, y) as input,\n",
" and returns a score\n",
" \n",
" Returns\n",
" -------\n",
" The average score obtained by randomly splitting (x, y) into training and \n",
" test sets, fitting on the training set, and evaluating score_func on the test set\n",
" \n",
" Examples\n",
" cv_score(clf, x, y, log_likelihood)\n",
" \"\"\"\n",
" result = 0\n",
" nfold = 5\n",
" for train, test in KFold(y.size, nfold): # split data into train/test groups, 5 times\n",
" clf.fit(x[train], y[train]) # fit\n",
" result += score_func(clf, x[test], y[test]) # evaluate score function on held-out data\n",
" return result / nfold # average"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 20
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"**3.7**\n",
"\n",
"Fill in the remaining code in this block, to loop over many values of `alpha` and `min_df` to determine\n",
"which settings are \"best\" in the sense of maximizing the cross-validated log-likelihood"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"#the grid of parameters to search over\n",
"alphas = [0, .1, 1, 5, 10, 50]\n",
"min_dfs = [1e-5, 1e-4, 1e-3, 1e-2, 1e-1]\n",
"\n",
"#Find the best value for alpha and min_df, and the best classifier\n",
"best_alpha = None\n",
"best_min_df = None\n",
"max_loglike = -np.inf\n",
"\n",
"for alpha in alphas:\n",
" for min_df in min_dfs: \n",
" vectorizer = CountVectorizer(min_df = min_df) \n",
" X, Y = make_xy(critics, vectorizer)\n",
" \n",
" #your code here\n",
" clf = MultinomialNB(alpha=alpha)\n",
" loglike = cv_score(clf, X, Y, log_likelihood)\n",
"\n",
" if loglike > max_loglike:\n",
" max_loglike = loglike\n",
" best_alpha, best_min_df = alpha, min_df"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 21
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"print \"alpha: %f\" % best_alpha\n",
"print \"min_df: %f\" % best_min_df"
],
"language": "python",
"metadata": {},
"outputs": [
{
"output_type": "stream",
"stream": "stdout",
"text": [
"alpha: 5.000000\n",
"min_df: 0.001000\n"
]
}
],
"prompt_number": 22
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"**3.8** Now that you've determined values for alpha and min_df that optimize the cross-validated log-likelihood, repeat the steps in 3.1, 3.2, and 3.4 to train a final classifier with these parameters, re-evaluate the accuracy, and draw a new calibration plot."
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"#Your code here\n",
"\n",
"vectorizer = CountVectorizer(min_df=best_min_df)\n",
"X, Y = make_xy(critics, vectorizer)\n",
"xtrain, xtest, ytrain, ytest = train_test_split(X, Y)\n",
"\n",
"clf = MultinomialNB(alpha=best_alpha).fit(xtrain, ytrain)\n",
"\n",
"calibration_plot(clf, xtest, ytest)\n",
"\n",
"# Your code here. Print the accuracy on the test and training dataset\n",
"training_accuracy = clf.score(xtrain, ytrain)\n",
"test_accuracy = clf.score(xtest, ytest)\n",
"\n",
"print \"Accuracy on training data: %0.2f\" % (training_accuracy)\n",
"print \"Accuracy on test data: %0.2f\" % (test_accuracy)"
],
"language": "python",
"metadata": {},
"outputs": [
{
"output_type": "stream",
"stream": "stdout",
"text": [
"Accuracy on training data: 0.78\n",
"Accuracy on test data: 0.74\n"
]
},
{
"metadata": {},
"output_type": "display_data",
"png": "iVBORw0KGgoAAAANSUhEUgAAAnYAAAGJCAYAAAAHeH6oAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
"text": [
""
]
}
],
"prompt_number": 23
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"**3.9** Discuss the various ways in which Cross-Validation has affected the model. Is the new model more or less accurate? Is overfitting better or worse? Is the model more or less calibrated?"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"**Answer** *The new model is slightly less accurate on the test data (74% vs 77% on the original model). However, it is both better calibrated and less over-fit than before. In other words, while the classification accuracy is slightly worse, the probabilities themselves are more accurate. The model is still slightly over-confident when making low P(Fresh) predictions. However, the calibration plot shows the model is usually within 1 error bar of the expected performance where P(Fresh) >= 0.2. Finally, the new model makes less-conclusive predictions on average -- the histogram in the calibration plot is more uniformly distributed, with fewer predictions clustered around P(Fresh) = 0 or 1.*"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"*To think about/play with, but not to hand in: What would happen if you tried this again using a function besides the log-likelihood -- for example, the classification accuracy?*"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Part 4: Interpretation. What words best predict a fresh or rotten review?\n",
"\n",
"**4.1**\n",
"Using your classifier and the `vectorizer.get_feature_names` method, determine which words best predict a positive or negative review. Print the 10 words\n",
"that best predict a \"fresh\" review, and the 10 words that best predict a \"rotten\" review. For each word, what is the model's probability of freshness if the word appears one time?\n",
"\n",
"#### Hints\n",
"\n",
"* Try computing the classification probability for a feature vector which consists of all 0s, except for a single 1. What does this probability refer to?\n",
"\n",
"* `np.eye` generates a matrix where the ith row is all 0s, except for the ith column which is 1."
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"## Your code here\n",
"\n",
"words = np.array(vectorizer.get_feature_names())\n",
"\n",
"x = np.eye(xtest.shape[1])\n",
"probs = clf.predict_log_proba(x)[:, 0]\n",
"ind = np.argsort(probs)\n",
"\n",
"good_words = words[ind[:10]]\n",
"bad_words = words[ind[-10:]]\n",
"\n",
"good_prob = probs[ind[:10]]\n",
"bad_prob = probs[ind[-10:]]\n",
"\n",
"print \"Good words\\t P(fresh | word)\"\n",
"for w, p in zip(good_words, good_prob):\n",
" print \"%20s\" % w, \"%0.2f\" % (1 - np.exp(p))\n",
" \n",
"print \"Bad words\\t P(fresh | word)\"\n",
"for w, p in zip(bad_words, bad_prob):\n",
" print \"%20s\" % w, \"%0.2f\" % (1 - np.exp(p))"
],
"language": "python",
"metadata": {},
"outputs": [
{
"output_type": "stream",
"stream": "stdout",
"text": [
"Good words\t P(fresh | word)\n",
" masterpiece 0.90\n",
" delight 0.89\n",
" remarkable 0.89\n",
" touching 0.87\n",
" perfect 0.87\n",
" rare 0.86\n",
" witty 0.86\n",
" superb 0.86\n",
" captures 0.86\n",
" delightful 0.85\n",
"Bad words\t P(fresh | word)\n",
" worst 0.21\n",
" dull 0.21\n",
" disappointment 0.20\n",
" disappointing 0.20\n",
" unfunny 0.20\n",
" bland 0.18\n",
" uninspired 0.17\n",
" pointless 0.17\n",
" lame 0.13\n",
" unfortunately 0.12\n"
]
}
],
"prompt_number": 22
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"**4.2**\n",
"\n",
"One of the best sources for inspiration when trying to improve a model is to look at examples where the model performs poorly. \n",
"\n",
"Find 5 fresh and rotten reviews where your model performs particularly poorly. Print each review."
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"#Your code here\n",
"x, y = make_xy(critics, vectorizer)\n",
"\n",
"prob = clf.predict_proba(x)[:, 0]\n",
"predict = clf.predict(x)\n",
"\n",
"bad_rotten = np.argsort(prob[y == 0])[:5]\n",
"bad_fresh = np.argsort(prob[y == 1])[-5:]\n",
"\n",
"print \"Mis-predicted Rotten quotes\"\n",
"print '---------------------------'\n",
"for row in bad_rotten:\n",
" print critics[y == 0].quote.irow(row)\n",
" print\n",
"\n",
"print \"Mis-predicted Fresh quotes\"\n",
"print '--------------------------'\n",
"for row in bad_fresh:\n",
" print critics[y == 1].quote.irow(row)\n",
" print"
],
"language": "python",
"metadata": {},
"outputs": [
{
"output_type": "stream",
"stream": "stdout",
"text": [
"Mis-predicted Rotten quotes\n",
"---------------------------\n",
"It survives today only as an unusually pure example of a typical 50s art-film strategy: the attempt to make the most modern and most popular of art forms acceptable to the intelligentsia by forcing it into an arcane, antique mold.\n",
"\n",
"The Waterboy is arguably Sandler's most enjoyable motion picture to date, but it's still far from a masterpiece.\n",
"\n",
"Pale Rider does nothing to disprove the wisdom that this genre is best left to the revival houses. A double feature of Shane and Eastwood's High Plains Drifter will do just fine, thanks.\n",
"\n",
"It's a sad day when an actor who's totally, beautifully in touch with his dark side finds himself stuck in a movie that's scared of its own shadow.\n",
"\n",
"Walken is one of the few undeniably charismatic male villains of recent years; he can generate a snakelike charm that makes his worst characters the most memorable, and here he operates on pure style.\n",
"\n",
"Mis-predicted Fresh quotes\n",
"--------------------------\n",
"The gangland plot is flimsy (bad guy Peter Greene wears too much eyeliner), and the jokes are erratic, but it's a far better showcase for Carrey's comic-from-Uranus talent than Ace Ventura.\n",
"\n",
"It's a one-joke movie, a funhouse ride, the cinematic equivalent of having a rubber spider thrown in your lap. But it doesn't matter if you reject the wispy script or the plot, which has as much substance as a spider's web; you'll jump every time.\n",
"\n",
"There's too much talent and too strong a story to mess it up. There was potential for more here, but this incarnation is nothing to be ashamed of, and some of the actors answer the bell.\n",
"\n",
"This tough-to-peg whodunit keeps you going for two hours, despite a few James Bond-ish (or Jane Bond-ish) turns that play less preposterously than you might assume were they to be divulged.\n",
"\n",
"Some of the gags don't work, but fewer than in any previous Brooks film that I've seen, and when the jokes are meant to be bad, they are riotously poor. What more can one ask of Mel Brooks?\n",
"\n"
]
}
],
"prompt_number": 23
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"**4.3** What do you notice about these mis-predictions? Naive Bayes classifiers assume that every word affects the probability independently of other words. In what way is this a bad assumption? In your answer, report your classifier's Freshness probability for the review \"This movie is not remarkable, touching, or superb in any way\"."
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"clf.predict_proba(vectorizer.transform(['This movie is not remarkable, touching, or superb in any way']))"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 25,
"text": [
"array([[ 0.01755082, 0.98244918]])"
]
}
],
"prompt_number": 25
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"**Answer** *Many mis-predictions seem due to the fact that the quotes use more ambivalent language -- quotes along the lines of \"this should have been a good movie, but it wasn't\". Words like \"but\", \"not\", etc. act to negate the sentiment of words. However, because Naive Bayes treats each word separately, it isn't able to capture these kind of word interactions. Because the quote \"this movie is not remarkable, touching, or superb in any way\" contains typically positive words like remarkabke/touching/superb, the classifier gives it P(Fresh)=0.98.*"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"**4.4**\n",
"If this was your final project, what are 3 things you would try in order to build a more effective review classifier? What other exploratory or explanatory visualizations do you think might be helpful?"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"There are many things worth trying. Some examples:\n",
"\n",
"1. You could try to build a NB model where the features are word pairs instead of words. This would be smart enough to realize that \"not good\" and \"so good\" mean very different things. This technique doesn't scale very well, since these features are much more sparse (and hence harder to detect repeatable patterns within).\n",
"2. You could try a model besides NB, that would allow for interactions between words -- for example, a Random Forest classifier.\n",
"3. You could consider adding supplemental features -- information about genre, director, cast, etc.\n",
"4. You could build a visualization that prints word reviews, and visually encodes each word with size or color to indicate how that word contributes to P(Fresh). For example, really bad words could show up as big and red, good words as big and green, common words as small and grey, etc."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### How to Submit\n",
"\n",
"To submit your homework, create a folder named lastname_firstinitial_hw3 and place this notebook file in the folder. Double check that this file is still called HW3.ipynb, and that it contains your solutions. Please do **not** include the critics.csv data file, if you created one. Compress the folder (please use .zip compression) and submit to the CS109 dropbox in the appropriate folder. If we cannot access your work because these directions are not followed correctly, we will not grade your work.\n"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"---\n",
"*css tweaks in this cell*\n",
""
]
}
],
"metadata": {}
}
]
}