{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "# Análise da actividade parlamentar das XIV Legislatura: Dezembro de 2020" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Introdução" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "**Este bloco de notas é uma actualização ao trabalho inicial que pode ser acedido pelos atalhos acima e que descreve de forma detalhada todas as opções, com visibilidade de todo o código, explicação dos algoritmos utilizados e todo o processo de tratamento e exploração de dados; cobre toda a legistlatura até ao fim de 2020, cobrindo assim a votação do Orçamento de Estado - a versão inicial foi feita antes dessas votações.**\n", "\n", "O posicionamento absoluto e relativo dos vários partidos políticos no Parlamento português tem sido motivo de interesse redobrado nos últimos anos. A eleição de deputados de partidos sem anterior presença parlamentar tem alimentado o debate cujas implicações ideológicas foram vísiveis de forma bastante prática na problemática em torno da escolha de lugares: partidos desagradados com o lugar atribuído (“Iniciativa Liberal Descontente Com Lugar Atribuído a Deputado No Parlamento - TSF” 2020), dificuldades gerais em termos de arrumação dos deputados (Renascença 2019), questões de ordem mais ou menos prática em torno de acessos (Almeida 2019), enfim, várias dimensões para uma questão que acaba por revelar a importância simbólica do posicionamento absoluto e relativo de cada partido no hemiciclo.\n", "\n", "Esta questão não é particularmente nova (Lourenço 2020), colocando-se em maior ou em menor grau com a entrada de novos partidos e a consequente necessidade de tomada de posição por parte do recém-chegado partido e a harmonização (possível) com os restantes, sendo que a sua posterior actividade parlamentar (nas suas diversas vertentes) poderá ou não alinhar-se com a sua auto-identificação (reflectida ou não nos lugares no hemiciclo).\n", "\n", "O ponto de partida para esta análise foi precisamente tentar descobrir se exclusivamente com base na actividade parlamentar, e em concreto no registo de votações, é possível estabelecer relações de proximidade e distância que permitam um agrupamento que não dependa de classificações a priori, e se sim, de que forma estes agrupamentos confirmam ou divergem da percepção existente?\n", "\n", "A utilização de dados abertos disponibilizados pelo Parlamento torna esta análise substancialmente mais simples, embora não sem a necessidade de tratamento e validação dos dados; de um ponto de vista prático este bloco de notas demonstra como aceder e transformar os dados de uma forma que pode ser útil para outras análises. No cenário nacional referência para a iniciativa http://hemiciclo.pt que, em linha com iniciativas europeias semelhantes, fornecesse um interface para um maior escrutinio da actividade parlamentar e um conjunto alargado de indicadores directos e indirectos do maior interesse (Sapage 2020). O presente trabalho tem alguns pontos de contacto com esta iniciativa, dentro dos limites que o seu objectivo pedagógico estabelece.\n", "\n", "A combinação de dados abertos com um bloco de notas Jupyter permite que o leitor tenha visibilidade dos vários passos e transformações (Randles et al. 2017), o que pode por vezes apresentar uma excessiva complexidade para quem não tenha familiaridade com programação; tentámos obviar esta limitação através da descrição das várias acções de forma a que se possa seguir a lógica e fruir dos resultados. Esta transparência assume uma dimensão adicional tendo em conta a temática que nos proposmos analisar, embora seja importante de forma tranversal (sobre a importância da repetibilidade, rastreabilidade, acesso e o papel de blocos Jupyter no contexto de open science ver, entre outros, exemplos em ecologia (Powers and Hampton 2019) astronomia (Wofford et al. 2019)).\n", "\n" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Metodologia" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Com base nos dados disponibilizados pela Assembleia da República em formato XML [DadosAbertos] são criadas _dataframes_ (tabelas de duas dimensões) com base na selecção de informação relativa aos padrões de votação de cada partido (e/ou deputados não-inscritos).\n", "\n", "São fundamentalmente feitas as seguintes análises:\n", "\n", "1. Vista geral das votações de cada partido, visualizado através de um _heatmap_\n", "2. Matriz de distância entre todos os partidos e dendograma\n", "3. Identificação de grupos (_spectral clustering_) e visualização das distâncias num espaço cartesiano (_multidimensional scaling_)\n", "\n", "\n", "O tratamento prévio dos dados em formato XML é feito de forma a seleccionar as votações de cada partido (ou deputado não inscrito); este processo tem alguma complexidade que se prende com o próprio processo de votação parlamentar, com múltiplas sessões e votações, pelo que foram \n", "\n", "De forma acessória são também feitas algumas análises adicionais, já mais removidas do objectivo central de determinação do distânciamento mas que complementam o quadro geral do que é possível." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Obtenção e tratamento dos dados\n", "\n", "Esta fase é fundamental para toda a restante análise: é onde obtemos os dados e os transformamos em informação num formato que pode ser facilmente manipulado." ] }, { "cell_type": "code", "execution_count": 41, "metadata": {}, "outputs": [], "source": [ "!pip3 install --user -q itables matplotlib pandas bs4 html5lib lxml seaborn sklearn pixiedust\n", "\n", "%matplotlib inline\n", "\n", "from itables import show\n", "import itables.options as opt\n", "\n", "opt.maxColumns=100\n", "opt.maxRows=2000\n", "opt.lengthMenu = [10, 20, 50, 100, 200, 500]\n" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Obtenção do ficheiro e conversão para dataframe" ] }, { "cell_type": "code", "execution_count": 3, "metadata": {}, "outputs": [], "source": [ "from urllib.request import urlopen\n", "import xml.etree.ElementTree as ET\n", "\n", "ini_url = 'http://app.parlamento.pt/webutils/docs/doc.xml?path=6148523063446f764c324679626d56304c3239775a57356b595852684c3052685a47397a51574a6c636e52766379394a626d6c6a6157463061585a68637939595356596c4d6a424d5a57647063327868644856795953394a626d6c6a6157463061585a686331684a566935346257773d&fich=IniciativasXIV.xml&Inline=true'\n", "ini_tree = ET.parse(urlopen(ini_url))" ] }, { "cell_type": "code", "execution_count": 4, "metadata": {}, "outputs": [], "source": [ "from bs4 import BeautifulSoup\n", "import re\n", "\n", "## Iteract through the existing dict\n", "def party_from_votes (votes):\n", " \"\"\"\n", " Determines the position of a party based on the majority position by summing all the individual votes.\n", " Argument is a dictionary returned by parse_voting()\n", " Returns a dictionary with the majority position of each party\n", " \"\"\"\n", " party_vote = {}\n", " for k, v in votes.items():\n", " ## Erase the name of the MP and keep the party only\n", " ## only when it's not from the \"Ninsc\" group - \n", " ## these need to be differentiated by name\n", " if re.match(\".*\\(Ninsc\\)\" , k) is None:\n", " nk = re.sub(r\".*\\((.+?)\\).*\", r\"\\1\", k)\n", " else:\n", " nk = k\n", " ## If it's the first entry for a key, create it\n", " if nk not in party_vote:\n", " party_vote[nk] = [0,0,0]\n", " ## Add to a specific index in a list\n", " if v == \"A Favor\":\n", " party_vote[nk][0] += 1\n", " elif v == \"Abstenção\":\n", " party_vote[nk][1] += 1\n", " elif v == \"Contra\":\n", " party_vote[nk][2] += 1\n", " for k,v in party_vote.items():\n", " party_vote[k]=[\"A Favor\", \"Abstenção\", \"Contra\"][v.index(max(v))]\n", " return party_vote\n", "\n", "def parse_voting(v_str):\n", " \"\"\"Parses the voting details in a string and returns a dict.\n", " \n", " Keyword arguments:\n", " \n", " v_str: a string with the description of the voting behaviour.\n", " \"\"\"\n", " ## Split by the HTML line break and put it in a dict\n", " d = dict(x.split(':') for x in v_str.split('
'))\n", " ## Remove the HTML tags\n", " for k, v in d.items():\n", " ctext = BeautifulSoup(v, \"lxml\")\n", " d[k] = ctext.get_text().strip().split(\",\")\n", " ## Invert the dict to get a 1-to-1 mapping\n", " ## and trim it\n", " votes = {}\n", " if len(v_str) < 1000: # Naive approach but realistically speaking... works well enough.\n", " for k, v in d.items():\n", " for p in v:\n", " if (p != ' ' and # Bypass empty entries\n", " re.match(\"[0-9]+\", p.strip()) is None and # Bypass quantified divergent voting patterns\n", " (re.match(\".*\\w +\\(.+\\)\", p.strip()) is None or # Bypass individual votes...\n", " re.match(\".*\\(Ninsc\\)\" , p.strip()) is not None)): # ... except when coming from \"Ninsc\"\n", " #print(\"|\"+ p.strip() + \"|\" + \":\\t\" + k)\n", " votes[p.strip()] = k\n", " else: # This is a nominal vote since the size of the string is greater than 1000\n", " for k, v in d.items():\n", " for p in v:\n", " if p != ' ':\n", " votes[p.strip()] = k\n", " ## Call the auxiliary function to produce the party position based on the majority votes\n", " votes = party_from_votes(votes)\n", " return votes" ] }, { "cell_type": "code", "execution_count": 5, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ ".........................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................2210\n" ] } ], "source": [ "import collections\n", "\n", "root = ini_tree\n", "\n", "counter=0\n", "\n", "## We will build a dataframe from a list of dicts\n", "## Inspired by the approach of Chris Moffitt here https://pbpython.com/pandas-list-dict.html\n", "init_list = []\n", "\n", "for voting in ini_tree.findall(\".//pt_gov_ar_objectos_VotacaoOut\"):\n", " votep = voting.find('./detalhe')\n", " if votep is not None:\n", " init_dict = collections.OrderedDict()\n", " counter +=1 \n", " init_dict['id'] = voting.find('id').text\n", " ## Add the \"I\" for Type to mark this as coming from \"Iniciativas\"\n", " init_dict['Tipo'] = \"I\"\n", " for c in voting:\n", " if c.tag == \"detalhe\":\n", " for party, vote in parse_voting(c.text).items():\n", " init_dict[party] = vote \n", " elif c.tag == \"descricao\":\n", " init_dict[c.tag] = c.text\n", " elif c.tag == \"ausencias\":\n", " init_dict[c.tag] = c.find(\"string\").text\n", " else:\n", " init_dict[c.tag] = c.text\n", " init_list.append(init_dict)\n", " ## Provide progression feedback\n", " print('.', end='')\n", " \n", "print(counter)" ] }, { "cell_type": "code", "execution_count": 44, "metadata": {}, "outputs": [], "source": [ "import pandas as pd\n", "\n", "ini_df = pd.DataFrame(init_list)\n", "#print(ini_df.shape)\n", "#ini_df.head()\n" ] }, { "cell_type": "code", "execution_count": 45, "metadata": {}, "outputs": [], "source": [ "## Copy Livre voting record to new aggregate columns...\n", "ini_df[\"L/JKM\"] = ini_df[\"L\"]\n", "## ... and fill the NAs with JKM voting record.\n", "#ini_df[\"L/JKM\"] = ini_df[\"L/JKM\"].fillna(ini_df[\"Joacine Katar Moreira (Ninsc)\"])\n", "#ini_df[[\"descricao\",\"L\",\"Joacine Katar Moreira (Ninsc)\",\"L/JKM\"]]" ] }, { "cell_type": "code", "execution_count": 46, "metadata": {}, "outputs": [], "source": [ "## Copy PAN voting record to new aggregate columns...\n", "ini_df[\"PAN/CR\"] = ini_df[\"PAN\"]\n", "## ... and update/replace with CR voting where it exists\n", "#ini_df[\"PAN/CR\"].update(ini_df[\"Cristina Rodrigues (Ninsc)\"])\n", "#ini_df[[\"descricao\",\"PAN\",\"Cristina Rodrigues (Ninsc)\",\"PAN/CR\"]]\n" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Actividades" ] }, { "cell_type": "code", "execution_count": 9, "metadata": {}, "outputs": [], "source": [ "act_url = 'http://app.parlamento.pt/webutils/docs/doc.xml?path=6148523063446f764c324679626d56304c3239775a57356b595852684c3052685a47397a51574a6c636e52766379394264476c32615752685a47567a4c31684a566955794d45786c5a326c7a6247463064584a684c30463061585a705a47466b5a584e595356597565473173&fich=AtividadesXIV.xml&Inline=true'\n", "act_tree = ET.parse(urlopen(act_url))" ] }, { "cell_type": "code", "execution_count": 10, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "............................................................................................................................................................................................................................................................................268\n" ] } ], "source": [ "import re\n", "import collections\n", "\n", "root = act_tree\n", "\n", "counter=0\n", "\n", "## We will build a dataframe from a list of dicts\n", "## Inspired by the approach of Chris Moffitt here https://pbpython.com/pandas-list-dict.html\n", "act_list = []\n", "\n", "def get_toplevel_desc (vid, tree):\n", " \"\"\"\n", " Gets the top-level title from a voting id\n", " \"\"\"\n", " for c in tree.find(\".//pt_gov_ar_objectos_VotacaoOut/[id='\"+ vid +\"']/../..\"):\n", " if c.tag == \"assunto\":\n", " return c.text\n", "\n", "for voting in act_tree.findall(\".//pt_gov_ar_objectos_VotacaoOut\"):\n", " act_dict = collections.OrderedDict()\n", " counter +=1\n", " votep = voting.find('./detalhe')\n", " if votep is not None:\n", " act_dict['id'] = voting.find('id').text\n", " ## Add the \"A\" for Type to mark this as coming from \"Iniciativas\"\n", " act_dict['Tipo'] = \"A\"\n", " for c in voting:\n", " if c.tag == \"id\":\n", " act_dict['descricao'] = get_toplevel_desc(c.text, act_tree)\n", " if c.tag == \"detalhe\":\n", " for party, vote in parse_voting(c.text).items():\n", " act_dict[party] = vote \n", " elif c.tag == \"ausencias\":\n", " act_dict[c.tag] = c.find(\"string\").text\n", " else:\n", " act_dict[c.tag] = c.text\n", " act_list.append(act_dict)\n", " ## Provide progression feedback\n", " print('.', end='')\n", "\n", "print(counter)" ] }, { "cell_type": "code", "execution_count": 48, "metadata": {}, "outputs": [], "source": [ "act_df = pd.DataFrame(act_list)\n", "#print(act_df.shape)\n", "\n", "#act_df.head()" ] }, { "cell_type": "code", "execution_count": 49, "metadata": {}, "outputs": [], "source": [ "## Copy Livre voting record to new aggregate columns...\n", "act_df[\"L/JKM\"] = act_df[\"L\"]\n", "## ... and fill the NAs with JKM voting record.\n", "act_df[\"L/JKM\"] = act_df[\"L/JKM\"].fillna(act_df[\"Joacine Katar Moreira (Ninsc)\"])\n", "\n", "## Copy PAN voting record to new aggregate columns...\n", "act_df[\"PAN/CR\"] = act_df[\"PAN\"]\n", "## ... and update/replace with CR voting where it exists\n", "act_df[\"PAN/CR\"].update(act_df[\"Cristina Rodrigues (Ninsc)\"])\n", "#act_df[[\"descricao\",\"PAN\",\"Cristina Rodrigues (Ninsc)\",\"PAN/CR\"]].head()" ] }, { "cell_type": "code", "execution_count": 13, "metadata": {}, "outputs": [], "source": [ "votes = pd.concat([ini_df.drop([\"tipoReuniao\"],axis=1),act_df.drop([\"data\",\"publicacao\"],axis=1)], sort=True)" ] }, { "cell_type": "code", "execution_count": 50, "metadata": {}, "outputs": [], "source": [ "votes_hm = votes[['BE', 'PCP', 'PEV', 'L/JKM', 'PS', 'PAN','PSD','IL','CDS-PP', 'CH']]\n", "#votes_hm.head()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Mapa térmico" ] }, { "cell_type": "code", "execution_count": 15, "metadata": {}, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "import matplotlib.pyplot as plt\n", "import matplotlib as mpl\n", "import seaborn as sns\n", "\n", "votes_hmn = votes_hm.replace([\"A Favor\", \"Contra\", \"Abstenção\", \"Ausência\"], [1,-1,0,2]).fillna(0)\n", "\n", "voting_palette = [\"#FB6962\",\"#FCFC99\",\"#79DE79\", \"black\"]\n", "\n", "fig = plt.figure(figsize=(8,8))\n", "sns.heatmap(votes_hmn,\n", " square=False,\n", " yticklabels = False,\n", " cbar=False,\n", " cmap=sns.color_palette(voting_palette),\n", " )\n", "plt.show()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Quem vota com quem\n", "\n", "Com estes dados podemos tentar obter uma resposta mais clara do que o \"mapa térmico\" anterior nos apresenta como sendo semelhanças e diferenças no registo de votação.\n", "\n", "Uma das questões que se coloca (e normalmente coloca-se com maior ênfase sempre que há uma votação que é apontada como sendo \"atípica\", com base na percepção geral do que é o comportamente de voto habitual de cada partido) é saber \"quem vota com quem\". Estes dados podem ser obtidos através da identificação, para cada partido, da quantidade de votações onde cada outro votou da mesma forma e criação de uma tabela com os resultados:" ] }, { "cell_type": "code", "execution_count": 16, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Total voting instances: 2457\n" ] }, { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
BEPCPPEVL/JKMPSPANPSDILCDS-PPCH
BE2457197320312047118918511121120411061110
PCP1973245723261819115715771093117611001093
PEV2031232624571879114316611066118410511069
L/JKM2047181918792457108417621037117410301040
PS118911571143108424571214164112201287912
PAN1851157716611762121424571248137312091171
PSD1121109310661037164112482457147117231344
IL1204117611841174122013731471245715611476
CDS-PP1106110010511030128712091723156124571525
CH111010931069104091211711344147615252457
\n", "
" ], "text/plain": [ " BE PCP PEV L/JKM PS PAN PSD IL CDS-PP CH\n", "BE 2457 1973 2031 2047 1189 1851 1121 1204 1106 1110\n", "PCP 1973 2457 2326 1819 1157 1577 1093 1176 1100 1093\n", "PEV 2031 2326 2457 1879 1143 1661 1066 1184 1051 1069\n", "L/JKM 2047 1819 1879 2457 1084 1762 1037 1174 1030 1040\n", "PS 1189 1157 1143 1084 2457 1214 1641 1220 1287 912\n", "PAN 1851 1577 1661 1762 1214 2457 1248 1373 1209 1171\n", "PSD 1121 1093 1066 1037 1641 1248 2457 1471 1723 1344\n", "IL 1204 1176 1184 1174 1220 1373 1471 2457 1561 1476\n", "CDS-PP 1106 1100 1051 1030 1287 1209 1723 1561 2457 1525\n", "CH 1110 1093 1069 1040 912 1171 1344 1476 1525 2457" ] }, "execution_count": 16, "metadata": {}, "output_type": "execute_result" } ], "source": [ "import numpy as np\n", "pv_list = []\n", "print(\"Total voting instances: \", votes_hm.shape[0])\n", "\n", "## Not necessarily the most straightforard way (check .crosstab or .pivot_table, possibly with pandas.melt and/or groupby)\n", "## but follows the same approach as before in using a list of dicts\n", "for party in votes_hm.columns:\n", " pv_dict = collections.OrderedDict()\n", " for column in votes_hmn:\n", " pv_dict[column]=votes_hmn[votes_hmn[party] == votes_hmn[column]].shape[0]\n", " pv_list.append(pv_dict)\n", "\n", "pv = pd.DataFrame(pv_list,index=votes_hm.columns)\n", "pv" ] }, { "cell_type": "code", "execution_count": 17, "metadata": {}, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "fig = plt.figure(figsize=(8,8))\n", "ax = fig.add_subplot()\n", "\n", "sns.heatmap(\n", " pv,\n", " cmap=sns.color_palette(\"mako_r\"),\n", " linewidth=1,\n", " annot = True,\n", " square =True,\n", " fmt=\"d\",\n", " cbar_kws={\"shrink\": 0.8})\n", "plt.title('Portuguese Parliament 14th Legislature, identical voting count')\n", "\n", "plt.show()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Matriz de distância\n", "\n", "Com base nos histórico de votações de cada partido produzimos uma matriz de distâncias entre eles; uma matriz de distâncias é uma matriz quadradra $n\\times n$ (onde _n_ é o número de partidos) e onde a distância entre _p_ e _q_ é o valor de $ d_{pq} $.\n", "\n", "$ \n", "\\begin{equation}\n", "D= \\begin{bmatrix} d_{11} & d_{12} & \\cdots & d_{1 n} \\\\ d_{21} & d_{22} & \\cdots & d_{2 n} \\\\ \\vdots & \\vdots & \\ddots & \\vdots \\\\ d_{31} & d_{32} & \\cdots & d_{n n} \\end{bmatrix}_{\\ n\\times n} \n", "\\end{equation}\n", "$\n", "\n", "\n", "A distância é obtida através da comparação de todas as observações de cada par usando uma determinada métrica de distância, sendo a distância euclideana bastante comum em termos gerais e também dentro de estudos sobre o mesmo domínio temático _(Krilavičius and Žilinskas 2008)_: cada elemento da matriz representa $ d\\left( p,q\\right) = \\sqrt {\\sum _{i=1}^{n} \\left( q_{i}-p_{i}\\right)^2 }$, equivalente, para dois pontos $P,Q $ , à mais genérica distância de Minkowski $ D\\left(P,Q\\right)=\\left(\\sum _{i=1}^{n}|x_{i}-y_{i}|^{p}\\right)^{\\frac {1}{p}} $ para $ p = 1$, mas note-se que a diagonal da matrix irá representar a distância entre um partido e ele próprio, logo $ d_{11} = d_{22} = \\dots = d_{nn} = 0 $.\n", "\n", "Na secção [Distâncias e matrizes](Distâncias_e_matrizes) colocámos uma discussão mais detalhada (mas passo-a-passo e destinada a quem não tenha necessariamente presente a matemática utilizada) sobre distâncias, _clustering_ e como são calculdadas, para quem tenha interesse numa compreensão mais quantitativa da matéria.\n", "\n", "A conversão de votos em representações númericas pode ser feita de várias formas _(Hix, Noury, and Roland 2006)_; adoptamos a abordagem de Krilavičius & Žilinskas (2008) no já citado trabalho relativo às votações no parlamento lituano por nos parecer apropriada à realidade portuguesa:\n", "\n", "* A favor: 1\n", "* Contra: -1\n", "* Abstenção: 0\n", "* Ausência: 0\n", "\n", "Este ponto é (mais um) dos que de forma relativamente opaca - pois raramente os detalhes têm a mesma projecção que os resultado finais - podem influenciar os resultados; cremos que em particular a equiparação entre _abstenção_ e _ausência_ merece alguma reflexão: considerámos que uma ausência em determinada votação tem um peso equivalente à abstenção, embora uma de forma passiva e outra activa.\n", "\n", "Para obtermos a matriz de distância usamos a função `pdist` e construímos um _dataframe_ que é uma matriz simétrica das distâncias entre os partidos." ] }, { "cell_type": "code", "execution_count": 20, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
BEPCPPEVL/JKMPSPANPSDILCDS-PPCH
BE0.00000029.00000025.57342422.69361163.55312733.89690361.34329652.49761956.87706049.386233
PCP29.0000000.00000015.06651931.11269860.23288142.04759258.18934652.50714254.42425948.249352
PEV25.57342415.0665190.00000028.23118861.50609739.50949359.64059052.74466856.11595149.183331
L/JKM22.69361131.11269828.2311880.00000062.91263833.55592360.01666450.60632455.73149948.062459
PS63.55312760.23288161.50609762.9126380.00000062.81719544.02272154.50688050.49752556.727418
PAN33.89690342.04759239.50949333.55592362.8171950.00000056.70978847.19110152.76362446.432747
PSD61.34329658.18934659.64059060.01666444.02272156.7097880.00000043.94314534.84250343.058100
IL52.49761952.50714252.74466850.60632454.50688047.19110143.9431450.00000036.56501137.054015
CDS-PP56.87706054.42425956.11595155.73149950.49752552.76362434.84250336.5650110.00000034.263683
CH49.38623348.24935249.18333148.06245956.72741846.43274743.05810037.05401534.2636830.000000
\n", "
" ], "text/plain": [ " BE PCP PEV L/JKM PS PAN \\\n", "BE 0.000000 29.000000 25.573424 22.693611 63.553127 33.896903 \n", "PCP 29.000000 0.000000 15.066519 31.112698 60.232881 42.047592 \n", "PEV 25.573424 15.066519 0.000000 28.231188 61.506097 39.509493 \n", "L/JKM 22.693611 31.112698 28.231188 0.000000 62.912638 33.555923 \n", "PS 63.553127 60.232881 61.506097 62.912638 0.000000 62.817195 \n", "PAN 33.896903 42.047592 39.509493 33.555923 62.817195 0.000000 \n", "PSD 61.343296 58.189346 59.640590 60.016664 44.022721 56.709788 \n", "IL 52.497619 52.507142 52.744668 50.606324 54.506880 47.191101 \n", "CDS-PP 56.877060 54.424259 56.115951 55.731499 50.497525 52.763624 \n", "CH 49.386233 48.249352 49.183331 48.062459 56.727418 46.432747 \n", "\n", " PSD IL CDS-PP CH \n", "BE 61.343296 52.497619 56.877060 49.386233 \n", "PCP 58.189346 52.507142 54.424259 48.249352 \n", "PEV 59.640590 52.744668 56.115951 49.183331 \n", "L/JKM 60.016664 50.606324 55.731499 48.062459 \n", "PS 44.022721 54.506880 50.497525 56.727418 \n", "PAN 56.709788 47.191101 52.763624 46.432747 \n", "PSD 0.000000 43.943145 34.842503 43.058100 \n", "IL 43.943145 0.000000 36.565011 37.054015 \n", "CDS-PP 34.842503 36.565011 0.000000 34.263683 \n", "CH 43.058100 37.054015 34.263683 0.000000 " ] }, "execution_count": 20, "metadata": {}, "output_type": "execute_result" } ], "source": [ "from scipy.spatial.distance import squareform\n", "from scipy.spatial.distance import pdist\n", "import scipy.spatial as sp, scipy.cluster.hierarchy as hc\n", "from itables import show\n", "\n", "votes_hmn = votes_hm.replace([\"A Favor\", \"Contra\", \"Abstenção\", \"Ausência\"], [1,-1,0,0]).fillna(0)\n", "\n", "## Transpose the dataframe used for the heatmap\n", "votes_t = votes_hmn.transpose()\n", "\n", "## Determine the Eucledian pairwise distance\n", "## (\"euclidean\" is actually the default option)\n", "pwdist = pdist(votes_t, metric='euclidean')\n", "\n", "## Create a square dataframe with the pairwise distances: the distance matrix\n", "distmat = pd.DataFrame(\n", " squareform(pwdist), # pass a symmetric distance matrix\n", " columns = votes_t.index,\n", " index = votes_t.index\n", ")\n", "#show(distmat, scrollY=\"200px\", scrollCollapse=True, paging=False)\n", "\n", "## Normalise by scaling between 0-1, using dataframe max value to keep the symmetry.\n", "## This is essentially a cosmetic step to \n", "#distmat=((distmat-distmat.min().min())/(distmat.max().max()-distmat.min().min()))*1\n", "distmat" ] }, { "cell_type": "code", "execution_count": 39, "metadata": {}, "outputs": [ { "data": { "application/javascript": [ "/* Put everything inside the global mpl namespace */\n", "window.mpl = {};\n", "\n", "\n", "mpl.get_websocket_type = function() {\n", " if (typeof(WebSocket) !== 'undefined') {\n", " return WebSocket;\n", " } else if (typeof(MozWebSocket) !== 'undefined') {\n", " return MozWebSocket;\n", " } else {\n", " alert('Your browser does not have WebSocket support. ' +\n", " 'Please try Chrome, Safari or Firefox ≥ 6. ' +\n", " 'Firefox 4 and 5 are also supported but you ' +\n", " 'have to enable WebSockets in about:config.');\n", " };\n", "}\n", "\n", "mpl.figure = function(figure_id, websocket, ondownload, parent_element) {\n", " this.id = figure_id;\n", "\n", " this.ws = websocket;\n", "\n", " this.supports_binary = (this.ws.binaryType != undefined);\n", "\n", " if (!this.supports_binary) {\n", " var warnings = document.getElementById(\"mpl-warnings\");\n", " if (warnings) {\n", " warnings.style.display = 'block';\n", " warnings.textContent = (\n", " \"This browser does not support binary websocket messages. \" +\n", " \"Performance may be slow.\");\n", " }\n", " }\n", "\n", " this.imageObj = new Image();\n", "\n", " this.context = undefined;\n", " this.message = undefined;\n", " this.canvas = undefined;\n", " this.rubberband_canvas = undefined;\n", " this.rubberband_context = undefined;\n", " this.format_dropdown = undefined;\n", "\n", " this.image_mode = 'full';\n", "\n", " this.root = $('
');\n", " this._root_extra_style(this.root)\n", " this.root.attr('style', 'display: inline-block');\n", "\n", " $(parent_element).append(this.root);\n", "\n", " this._init_header(this);\n", " this._init_canvas(this);\n", " this._init_toolbar(this);\n", "\n", " var fig = this;\n", "\n", " this.waiting = false;\n", "\n", " this.ws.onopen = function () {\n", " fig.send_message(\"supports_binary\", {value: fig.supports_binary});\n", " fig.send_message(\"send_image_mode\", {});\n", " if (mpl.ratio != 1) {\n", " fig.send_message(\"set_dpi_ratio\", {'dpi_ratio': mpl.ratio});\n", " }\n", " fig.send_message(\"refresh\", {});\n", " }\n", "\n", " this.imageObj.onload = function() {\n", " if (fig.image_mode == 'full') {\n", " // Full images could contain transparency (where diff images\n", " // almost always do), so we need to clear the canvas so that\n", " // there is no ghosting.\n", " fig.context.clearRect(0, 0, fig.canvas.width, fig.canvas.height);\n", " }\n", " fig.context.drawImage(fig.imageObj, 0, 0);\n", " };\n", "\n", " this.imageObj.onunload = function() {\n", " fig.ws.close();\n", " }\n", "\n", " this.ws.onmessage = this._make_on_message_function(this);\n", "\n", " this.ondownload = ondownload;\n", "}\n", "\n", "mpl.figure.prototype._init_header = function() {\n", " var titlebar = $(\n", " '
');\n", " var titletext = $(\n", " '
');\n", " titlebar.append(titletext)\n", " this.root.append(titlebar);\n", " this.header = titletext[0];\n", "}\n", "\n", "\n", "\n", "mpl.figure.prototype._canvas_extra_style = function(canvas_div) {\n", "\n", "}\n", "\n", "\n", "mpl.figure.prototype._root_extra_style = function(canvas_div) {\n", "\n", "}\n", "\n", "mpl.figure.prototype._init_canvas = function() {\n", " var fig = this;\n", "\n", " var canvas_div = $('
');\n", "\n", " canvas_div.attr('style', 'position: relative; clear: both; outline: 0');\n", "\n", " function canvas_keyboard_event(event) {\n", " return fig.key_event(event, event['data']);\n", " }\n", "\n", " canvas_div.keydown('key_press', canvas_keyboard_event);\n", " canvas_div.keyup('key_release', canvas_keyboard_event);\n", " this.canvas_div = canvas_div\n", " this._canvas_extra_style(canvas_div)\n", " this.root.append(canvas_div);\n", "\n", " var canvas = $('');\n", " canvas.addClass('mpl-canvas');\n", " canvas.attr('style', \"left: 0; top: 0; z-index: 0; outline: 0\")\n", "\n", " this.canvas = canvas[0];\n", " this.context = canvas[0].getContext(\"2d\");\n", "\n", " var backingStore = this.context.backingStorePixelRatio ||\n", "\tthis.context.webkitBackingStorePixelRatio ||\n", "\tthis.context.mozBackingStorePixelRatio ||\n", "\tthis.context.msBackingStorePixelRatio ||\n", "\tthis.context.oBackingStorePixelRatio ||\n", "\tthis.context.backingStorePixelRatio || 1;\n", "\n", " mpl.ratio = (window.devicePixelRatio || 1) / backingStore;\n", "\n", " var rubberband = $('');\n", " rubberband.attr('style', \"position: absolute; left: 0; top: 0; z-index: 1;\")\n", "\n", " var pass_mouse_events = true;\n", "\n", " canvas_div.resizable({\n", " start: function(event, ui) {\n", " pass_mouse_events = false;\n", " },\n", " resize: function(event, ui) {\n", " fig.request_resize(ui.size.width, ui.size.height);\n", " },\n", " stop: function(event, ui) {\n", " pass_mouse_events = true;\n", " fig.request_resize(ui.size.width, ui.size.height);\n", " },\n", " });\n", "\n", " function mouse_event_fn(event) {\n", " if (pass_mouse_events)\n", " return fig.mouse_event(event, event['data']);\n", " }\n", "\n", " rubberband.mousedown('button_press', mouse_event_fn);\n", " rubberband.mouseup('button_release', mouse_event_fn);\n", " // Throttle sequential mouse events to 1 every 20ms.\n", " rubberband.mousemove('motion_notify', mouse_event_fn);\n", "\n", " rubberband.mouseenter('figure_enter', mouse_event_fn);\n", " rubberband.mouseleave('figure_leave', mouse_event_fn);\n", "\n", " canvas_div.on(\"wheel\", function (event) {\n", " event = event.originalEvent;\n", " event['data'] = 'scroll'\n", " if (event.deltaY < 0) {\n", " event.step = 1;\n", " } else {\n", " event.step = -1;\n", " }\n", " mouse_event_fn(event);\n", " });\n", "\n", " canvas_div.append(canvas);\n", " canvas_div.append(rubberband);\n", "\n", " this.rubberband = rubberband;\n", " this.rubberband_canvas = rubberband[0];\n", " this.rubberband_context = rubberband[0].getContext(\"2d\");\n", " this.rubberband_context.strokeStyle = \"#000000\";\n", "\n", " this._resize_canvas = function(width, height) {\n", " // Keep the size of the canvas, canvas container, and rubber band\n", " // canvas in synch.\n", " canvas_div.css('width', width)\n", " canvas_div.css('height', height)\n", "\n", " canvas.attr('width', width * mpl.ratio);\n", " canvas.attr('height', height * mpl.ratio);\n", " canvas.attr('style', 'width: ' + width + 'px; height: ' + height + 'px;');\n", "\n", " rubberband.attr('width', width);\n", " rubberband.attr('height', height);\n", " }\n", "\n", " // Set the figure to an initial 600x600px, this will subsequently be updated\n", " // upon first draw.\n", " this._resize_canvas(600, 600);\n", "\n", " // Disable right mouse context menu.\n", " $(this.rubberband_canvas).bind(\"contextmenu\",function(e){\n", " return false;\n", " });\n", "\n", " function set_focus () {\n", " canvas.focus();\n", " canvas_div.focus();\n", " }\n", "\n", " window.setTimeout(set_focus, 100);\n", "}\n", "\n", "mpl.figure.prototype._init_toolbar = function() {\n", " var fig = this;\n", "\n", " var nav_element = $('
');\n", " nav_element.attr('style', 'width: 100%');\n", " this.root.append(nav_element);\n", "\n", " // Define a callback function for later on.\n", " function toolbar_event(event) {\n", " return fig.toolbar_button_onclick(event['data']);\n", " }\n", " function toolbar_mouse_event(event) {\n", " return fig.toolbar_button_onmouseover(event['data']);\n", " }\n", "\n", " for(var toolbar_ind in mpl.toolbar_items) {\n", " var name = mpl.toolbar_items[toolbar_ind][0];\n", " var tooltip = mpl.toolbar_items[toolbar_ind][1];\n", " var image = mpl.toolbar_items[toolbar_ind][2];\n", " var method_name = mpl.toolbar_items[toolbar_ind][3];\n", "\n", " if (!name) {\n", " // put a spacer in here.\n", " continue;\n", " }\n", " var button = $('');\n", " button.click(method_name, toolbar_event);\n", " button.mouseover(tooltip, toolbar_mouse_event);\n", " nav_element.append(button);\n", " }\n", "\n", " // Add the status bar.\n", " var status_bar = $('');\n", " nav_element.append(status_bar);\n", " this.message = status_bar[0];\n", "\n", " // Add the close button to the window.\n", " var buttongrp = $('
');\n", " var button = $('');\n", " button.click(function (evt) { fig.handle_close(fig, {}); } );\n", " button.mouseover('Stop Interaction', toolbar_mouse_event);\n", " buttongrp.append(button);\n", " var titlebar = this.root.find($('.ui-dialog-titlebar'));\n", " titlebar.prepend(buttongrp);\n", "}\n", "\n", "mpl.figure.prototype._root_extra_style = function(el){\n", " var fig = this\n", " el.on(\"remove\", function(){\n", "\tfig.close_ws(fig, {});\n", " });\n", "}\n", "\n", "mpl.figure.prototype._canvas_extra_style = function(el){\n", " // this is important to make the div 'focusable\n", " el.attr('tabindex', 0)\n", " // reach out to IPython and tell the keyboard manager to turn it's self\n", " // off when our div gets focus\n", "\n", " // location in version 3\n", " if (IPython.notebook.keyboard_manager) {\n", " IPython.notebook.keyboard_manager.register_events(el);\n", " }\n", " else {\n", " // location in version 2\n", " IPython.keyboard_manager.register_events(el);\n", " }\n", "\n", "}\n", "\n", "mpl.figure.prototype._key_event_extra = function(event, name) {\n", " var manager = IPython.notebook.keyboard_manager;\n", " if (!manager)\n", " manager = IPython.keyboard_manager;\n", "\n", " // Check for shift+enter\n", " if (event.shiftKey && event.which == 13) {\n", " this.canvas_div.blur();\n", " // select the cell after this one\n", " var index = IPython.notebook.find_cell_index(this.cell_info[0]);\n", " IPython.notebook.select(index + 1);\n", " }\n", "}\n", "\n", "mpl.figure.prototype.handle_save = function(fig, msg) {\n", " fig.ondownload(fig, null);\n", "}\n", "\n", "\n", "mpl.find_output_cell = function(html_output) {\n", " // Return the cell and output element which can be found *uniquely* in the notebook.\n", " // Note - this is a bit hacky, but it is done because the \"notebook_saving.Notebook\"\n", " // IPython event is triggered only after the cells have been serialised, which for\n", " // our purposes (turning an active figure into a static one), is too late.\n", " var cells = IPython.notebook.get_cells();\n", " var ncells = cells.length;\n", " for (var i=0; i= 3 moved mimebundle to data attribute of output\n", " data = data.data;\n", " }\n", " if (data['text/html'] == html_output) {\n", " return [cell, data, j];\n", " }\n", " }\n", " }\n", " }\n", "}\n", "\n", "// Register the function which deals with the matplotlib target/channel.\n", "// The kernel may be null if the page has been refreshed.\n", "if (IPython.notebook.kernel != null) {\n", " IPython.notebook.kernel.comm_manager.register_target('matplotlib', mpl.mpl_figure_comm);\n", "}\n" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "text/html": [ "
" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "## Display the heatmap of the distance matrix\n", "\n", "fig = plt.figure(figsize=(8,8))\n", "ax = fig.add_subplot()\n", "\n", "sns.heatmap(\n", " distmat,\n", " cmap=sns.color_palette(\"Reds_r\"),\n", " linewidth=1,\n", " annot = True,\n", " square =True,\n", " cbar_kws={\"shrink\": 0.8})\n", "plt.title('Portuguese Parliament 14th Legislature, Distance Matrix')\n", "\n", "plt.show()" ] }, { "cell_type": "code", "execution_count": 37, "metadata": {}, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "## Perform hierarchical linkage on the distance matrix using Ward's method.\n", "distmat_link = hc.linkage(pwdist, method=\"ward\", optimal_ordering=True )\n", "\n", "sns.clustermap(\n", " distmat,\n", " annot = True,\n", " cmap=sns.color_palette(\"Reds_r\"),\n", " linewidth=1,\n", " #standard_scale=1,\n", " row_linkage=distmat_link,\n", " col_linkage=distmat_link,\n", " figsize=(8,8)).fig.suptitle('Portuguese Parliament 14th Legislature, Clustermap')\n", "\n", "plt.show()" ] }, { "cell_type": "code", "execution_count": 36, "metadata": {}, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "from scipy.cluster.hierarchy import dendrogram\n", "fig = plt.figure(figsize=(8,5))\n", "dendrogram(distmat_link, labels=votes_hmn.columns)\n", "\n", "plt.title(\"Portuguese Parliament 14th Legislature, Dendogram\")\n", "plt.show()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## _Clustering_ de observações: DBSCAN e _Spectrum Scaling_\n" ] }, { "cell_type": "code", "execution_count": 24, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
BEPCPPEVL/JKMPSPANPSDILCDS-PPCH
BE0.0000000.4563110.4023940.3570811.0000000.5333630.9652290.8260430.8949530.777086
PCP0.4563110.0000000.2370700.4895540.9477560.6616130.9156020.8261930.8563580.759197
PEV0.4023940.2370700.0000000.4442140.9677900.6216770.9384370.8299300.8829770.773893
L/JKM0.3570810.4895540.4442140.0000000.9899220.5279980.9443540.7962840.8769280.756256
PS1.0000000.9477560.9677900.9899220.0000000.9884200.6926920.8576590.7945720.892598
PAN0.5333630.6616130.6216770.5279980.9884200.0000000.8923210.7425460.8302290.730613
PSD0.9652290.9156020.9384370.9443540.6926920.8923210.0000000.6914400.5482420.677513
IL0.8260430.8261930.8299300.7962840.8576590.7425460.6914400.0000000.5753460.583040
CDS-PP0.8949530.8563580.8829770.8769280.7945720.8302290.5482420.5753460.0000000.539134
CH0.7770860.7591970.7738930.7562560.8925980.7306130.6775130.5830400.5391340.000000
\n", "
" ], "text/plain": [ " BE PCP PEV L/JKM PS PAN PSD \\\n", "BE 0.000000 0.456311 0.402394 0.357081 1.000000 0.533363 0.965229 \n", "PCP 0.456311 0.000000 0.237070 0.489554 0.947756 0.661613 0.915602 \n", "PEV 0.402394 0.237070 0.000000 0.444214 0.967790 0.621677 0.938437 \n", "L/JKM 0.357081 0.489554 0.444214 0.000000 0.989922 0.527998 0.944354 \n", "PS 1.000000 0.947756 0.967790 0.989922 0.000000 0.988420 0.692692 \n", "PAN 0.533363 0.661613 0.621677 0.527998 0.988420 0.000000 0.892321 \n", "PSD 0.965229 0.915602 0.938437 0.944354 0.692692 0.892321 0.000000 \n", "IL 0.826043 0.826193 0.829930 0.796284 0.857659 0.742546 0.691440 \n", "CDS-PP 0.894953 0.856358 0.882977 0.876928 0.794572 0.830229 0.548242 \n", "CH 0.777086 0.759197 0.773893 0.756256 0.892598 0.730613 0.677513 \n", "\n", " IL CDS-PP CH \n", "BE 0.826043 0.894953 0.777086 \n", "PCP 0.826193 0.856358 0.759197 \n", "PEV 0.829930 0.882977 0.773893 \n", "L/JKM 0.796284 0.876928 0.756256 \n", "PS 0.857659 0.794572 0.892598 \n", "PAN 0.742546 0.830229 0.730613 \n", "PSD 0.691440 0.548242 0.677513 \n", "IL 0.000000 0.575346 0.583040 \n", "CDS-PP 0.575346 0.000000 0.539134 \n", "CH 0.583040 0.539134 0.000000 " ] }, "execution_count": 24, "metadata": {}, "output_type": "execute_result" } ], "source": [ "import numpy as np\n", "\n", "distmat_mm=((distmat-distmat.min().min())/(distmat.max().max()-distmat.min().min()))*1\n", "pd.DataFrame(distmat_mm, distmat.index, distmat.columns)" ] }, { "cell_type": "code", "execution_count": 25, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
BEPCPPEVL/JKMPSPANPSDILCDS-PPCH
BE1.0000000.5436890.5976060.6429190.0000000.4666370.0347710.1739570.1050470.222914
PCP0.5436891.0000000.7629300.5104460.0522440.3383870.0843980.1738070.1436420.240803
PEV0.5976060.7629301.0000000.5557860.0322100.3783230.0615630.1700700.1170230.226107
L/JKM0.6429190.5104460.5557861.0000000.0100780.4720020.0556460.2037160.1230720.243744
PS0.0000000.0522440.0322100.0100781.0000000.0115800.3073080.1423410.2054280.107402
PAN0.4666370.3383870.3783230.4720020.0115801.0000000.1076790.2574540.1697710.269387
PSD0.0347710.0843980.0615630.0556460.3073080.1076791.0000000.3085600.4517580.322487
IL0.1739570.1738070.1700700.2037160.1423410.2574540.3085601.0000000.4246540.416960
CDS-PP0.1050470.1436420.1170230.1230720.2054280.1697710.4517580.4246541.0000000.460866
CH0.2229140.2408030.2261070.2437440.1074020.2693870.3224870.4169600.4608661.000000
\n", "
" ], "text/plain": [ " BE PCP PEV L/JKM PS PAN PSD \\\n", "BE 1.000000 0.543689 0.597606 0.642919 0.000000 0.466637 0.034771 \n", "PCP 0.543689 1.000000 0.762930 0.510446 0.052244 0.338387 0.084398 \n", "PEV 0.597606 0.762930 1.000000 0.555786 0.032210 0.378323 0.061563 \n", "L/JKM 0.642919 0.510446 0.555786 1.000000 0.010078 0.472002 0.055646 \n", "PS 0.000000 0.052244 0.032210 0.010078 1.000000 0.011580 0.307308 \n", "PAN 0.466637 0.338387 0.378323 0.472002 0.011580 1.000000 0.107679 \n", "PSD 0.034771 0.084398 0.061563 0.055646 0.307308 0.107679 1.000000 \n", "IL 0.173957 0.173807 0.170070 0.203716 0.142341 0.257454 0.308560 \n", "CDS-PP 0.105047 0.143642 0.117023 0.123072 0.205428 0.169771 0.451758 \n", "CH 0.222914 0.240803 0.226107 0.243744 0.107402 0.269387 0.322487 \n", "\n", " IL CDS-PP CH \n", "BE 0.173957 0.105047 0.222914 \n", "PCP 0.173807 0.143642 0.240803 \n", "PEV 0.170070 0.117023 0.226107 \n", "L/JKM 0.203716 0.123072 0.243744 \n", "PS 0.142341 0.205428 0.107402 \n", "PAN 0.257454 0.169771 0.269387 \n", "PSD 0.308560 0.451758 0.322487 \n", "IL 1.000000 0.424654 0.416960 \n", "CDS-PP 0.424654 1.000000 0.460866 \n", "CH 0.416960 0.460866 1.000000 " ] }, "execution_count": 25, "metadata": {}, "output_type": "execute_result" } ], "source": [ "affinmat_mm = pd.DataFrame(1-distmat_mm, distmat.index, distmat.columns)\n", "affinmat_mm " ] }, { "cell_type": "code", "execution_count": 26, "metadata": {}, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "sns.set(style=\"white\")\n", "\n", "## Make the top triangle\n", "mask = np.triu(np.ones_like(affinmat_mm, dtype=np.bool))\n", "fig = plt.figure(figsize=(8,8))\n", "ax = fig.add_subplot()\n", "plt.title('Portuguese Parliament 14th Legislature, Affinity Matrix')\n", "\n", "## Display the heatmap of the affinity matrix, masking the top triangle\n", "\n", "sns.heatmap(\n", " affinmat_mm,\n", " cmap=sns.color_palette(\"Greens\"),\n", " linewidth=1,\n", " annot = False,\n", " square =True,\n", " cbar_kws={\"shrink\": .8},\n", " mask=mask,linewidths=.5)\n", "\n", "plt.show()" ] }, { "cell_type": "code", "execution_count": 27, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "{'BE': 0,\n", " 'PCP': 0,\n", " 'PEV': 0,\n", " 'L/JKM': 0,\n", " 'PS': 1,\n", " 'PAN': 0,\n", " 'PSD': 1,\n", " 'IL': 1,\n", " 'CDS-PP': 1,\n", " 'CH': 1}" ] }, "execution_count": 27, "metadata": {}, "output_type": "execute_result" } ], "source": [ "from sklearn.cluster import DBSCAN\n", "\n", "dbscan_labels = DBSCAN(eps=1.1).fit(affinmat_mm)\n", "dbscan_labels.labels_\n", "dbscan_dict = dict(zip(distmat_mm,dbscan_labels.labels_))\n", "dbscan_dict" ] }, { "cell_type": "code", "execution_count": 28, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "{'BE': 2, 'PCP': 3, 'PEV': 3, 'L/JKM': 2, 'PS': 1, 'PAN': 2, 'PSD': 1, 'IL': 0, 'CDS-PP': 0, 'CH': 0}\n" ] } ], "source": [ "from sklearn.cluster import SpectralClustering\n", "sc = SpectralClustering(4, affinity=\"precomputed\",random_state=2020).fit_predict(affinmat_mm)\n", "sc_dict = dict(zip(distmat,sc))\n", "\n", "print(sc_dict)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### _Multidimensional scaling_ \n", "\n", "Até agora temos conseguido extrair informação interessante dos dados de votação:\n", "\n", "1. O mapa térmico de votação permite-nos uma primeira visão do comportamente de todos os partidos. \n", "2. A matriz de distâncias fornece-nos uma forma de comparar as distâncias entre os diferentes partidos através de um mapa térmico.\n", "3. O dendograma identifica de forma hierárquica agrupamentos.\n", "4. Através de DBSCAN e _Spectrum Clustering_ identificamos \"blocos\" com base na matriz de afinidade.\n", "\n", "Não temos ainda uma forma de visualizar a distância relativa de cada partido em relação aos outros com base nas distâncias/semelhanças: temos algo próximo com base no dendograma mas existem outras formas de visualização interessantes.\n", "\n", "Uma das formas é o _multidimensional scaling_ que permite visualizar a distância ao projectar em 2 ou 3 dimensões (também conhecidas como _dimensões visualizavies_) conjuntos multidimensionais, mantendo a distância relativa _(“Graphical Representation of Proximity Measures for Multidimensional Data « The Mathematica Journal” 2020)_.\n", "\n", "Como é habitual temos em Python, através da biblioteca `scikit-learn` (que já usámos para DBSCAN e _Spectrum Clustering_), uma implementação que podemos usar sem grande dificuldade _(“2.2. Manifold Learning — Scikit-Learn 0.23.2 Documentation” 2020)_." ] }, { "cell_type": "code", "execution_count": 29, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "array([[ 0.33299885, 0.36138221],\n", " [-0.00657669, 0.42473032],\n", " [ 0.10930041, 0.44054896],\n", " [ 0.39557388, 0.21339868],\n", " [-0.68909687, 0.15393775],\n", " [ 0.49318781, -0.08163115],\n", " [-0.51275165, -0.28664118],\n", " [ 0.14351589, -0.50188438],\n", " [-0.22776952, -0.48473615],\n", " [-0.03838211, -0.23910504]])" ] }, "execution_count": 29, "metadata": {}, "output_type": "execute_result" } ], "source": [ "from sklearn.manifold import MDS\n", "\n", "mds = MDS(n_components=2, dissimilarity='precomputed',random_state=2020, n_init=100, max_iter=1000)\n", "\n", "## We use the normalised distance matrix but results would\n", "## be similar with the original one, just with a different scale/axis\n", "results = mds.fit(distmat_mm.values)\n", "coords = results.embedding_\n", "coords" ] }, { "cell_type": "code", "execution_count": 30, "metadata": {}, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "## Graphic options\n", "sns.set()\n", "sns.set_style(\"ticks\")\n", "\n", "fig, ax = plt.subplots(figsize=(8,8))\n", "\n", "plt.title('Portuguese Parliament Voting Records Analysis, 14th Legislature', fontsize=14)\n", "\n", "for label, x, y in zip(distmat_mm.columns, coords[:, 0], coords[:, 1]):\n", " ax.scatter(x, y, s=250)\n", " ax.axis('equal')\n", " ax.annotate(label,xy = (x-0.02, y+0.025))\n", "plt.show()" ] }, { "cell_type": "code", "execution_count": 31, "metadata": {}, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "from sklearn.manifold import MDS\n", "import random\n", "\n", "sns.set()\n", "sns.set_style(\"ticks\")\n", "\n", "\n", "fig, ax = plt.subplots(figsize=(8,8))\n", "\n", "fig.suptitle('Portuguese Parliament Voting Records Analysis, 14th Legislature', fontsize=14)\n", "ax.set_title('MDS with DBSCAN clusters (2D)')\n", "\n", "for label, x, y in zip(distmat_mm.columns, coords[:, 0], coords[:, 1]):\n", " ax.scatter(x, y, c = \"C\"+str(dbscan_dict[label]), s=250)\n", " ax.axis('equal')\n", " ax.annotate(label,xy = (x-0.02, y+0.025))\n", "\n", "plt.show()" ] }, { "cell_type": "code", "execution_count": 32, "metadata": {}, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "from sklearn.manifold import MDS\n", "import random\n", "\n", "sns.set()\n", "sns.set_style(\"ticks\")\n", "\n", "fig, ax = plt.subplots(figsize=(8,8))\n", "fig.suptitle('Portuguese Parliament Voting Records Analysis, 14th Legislature', fontsize=14)\n", "ax.set_title('MDS with Spectrum Scaling clusters (2D)')\n", "\n", "\n", "for label, x, y in zip(distmat_mm.columns, coords[:, 0], coords[:, 1]):\n", " ax.scatter(x, y, c = \"C\"+str(sc_dict[label]), s=250)\n", " ax.axis('equal')\n", " ax.annotate(label,xy = (x-0.02, y+0.025))\n", "\n", "plt.show()" ] }, { "cell_type": "code", "execution_count": 33, "metadata": {}, "outputs": [], "source": [ "## From https://stackoverflow.com/questions/10374930/matplotlib-annotating-a-3d-scatter-plot\n", "\n", "from mpl_toolkits.mplot3d.proj3d import proj_transform\n", "from matplotlib.text import Annotation\n", "\n", "class Annotation3D(Annotation):\n", " '''Annotate the point xyz with text s'''\n", "\n", " def __init__(self, s, xyz, *args, **kwargs):\n", " Annotation.__init__(self,s, xy=(0,0), *args, **kwargs)\n", " self._verts3d = xyz \n", "\n", " def draw(self, renderer):\n", " xs3d, ys3d, zs3d = self._verts3d\n", " xs, ys, zs = proj_transform(xs3d, ys3d, zs3d, renderer.M)\n", " self.xy=(xs,ys)\n", " Annotation.draw(self, renderer)\n", " \n", "def annotate3D(ax, s, *args, **kwargs):\n", " '''add anotation text s to to Axes3d ax'''\n", "\n", " tag = Annotation3D(s, *args, **kwargs)\n", " ax.add_artist(tag)" ] }, { "cell_type": "code", "execution_count": 34, "metadata": {}, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "from sklearn.manifold import MDS\n", "import mpl_toolkits.mplot3d\n", "import random\n", "\n", "mds = MDS(n_components=3, dissimilarity='precomputed',random_state=1234, n_init=100, max_iter=1000)\n", "results = mds.fit(distmat.values)\n", "parties = distmat.columns\n", "coords = results.embedding_\n", "\n", "sns.set()\n", "sns.set_style(\"ticks\")\n", "\n", "fig = plt.figure(figsize=(8,8))\n", "ax = fig.add_subplot(111, projection='3d')\n", "\n", "fig.suptitle('Portuguese Parliament Voting Records Analysis, 14th Legislature', fontsize=14)\n", "ax.set_title('MDS with Spectrum Scaling clusters (3D)')\n", "\n", "for label, x, y, z in zip(parties, coords[:, 0], coords[:, 1], coords[:, 2]):\n", " #print(label,pmds_colors[label])\n", " ax.scatter(x, y, z, c=\"C\"+str(sc_dict[label]),s=250)\n", " annotate3D(ax, s=str(label), xyz=[x,y,z], fontsize=10, xytext=(-3,3),\n", " textcoords='offset points', ha='right',va='bottom') \n", "plt.show()\n" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [] } ], "metadata": { "kernelspec": { "display_name": "Python 3", "language": "python", "name": "python3" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.6.8" } }, "nbformat": 4, "nbformat_minor": 4 }