K-Means Clustering Output

After merging order_id and user_id:

1. 
   
   

Load the products csv file data:

1. 
   
   

Load the aisles csv file data:

1. 
   
   

Print shape of aisles:

1. 
   
   

Merge the aisle_id,product_id,order_id rows of products, orders and aisles:

1. 
   
   

Top 10 rows in the products list:

1. 
   
   

Length of total number of unique products list:

1. 
   
   

Shape of prior data:

1. 
   
   

Length of total number of unique aisles data list:

1. 
   
   

Fresh fruits and fresh vegetables are best selling top two products in the products list:

1. 
   
   

Perform cross tabulation or contingency tables of customer products with user_id and aisle_id:

1. 
   
   

Shape of customer products:

1. 
   
   

Perform Principal component analysis (PCA) on customer products from sklearn.decomposition import PCA:

1. 
   
   

Plot cluster of class 1 from matplotlib import pyplot as plt:

1. 
   
   

Provide random centers for each clusters:

1. 
   
   

Predict the one hundred fifty data points for generating clusters:

1. 
   
   

Plot clusters belonging to class 1, class 2,class 3 and class 4:

1. 
   
   

Plot sub plots of each clusters belonging to different classes :

1. 
   
   

Top 10 products belonging to cluster 0:

1. 
   
   

Top 10 products belonging to cluster 1:

1. 
   
   

Top 10 products belonging to cluster 2:

1. 
   
   

Top 10 products belonging to cluster 3:

1. 
   
   

Top 10 products belonging to cluster 4:

1. 
   
   

A first analysis of the clusters confirm the initial hypothesis that are products which are genereically bought by the majority of the customers:

• fresh fruits
• fresh vegetables
• yogurt
• packaged cheese
• milk

Ratio of purchasing orders of most frequently bought products:

1. 
   
   

Percentage of purchasing orders of most frequently bought products:

1.