

In [ ]:

    
%matplotlib inline
import pandas as pd

from recommend import recommend_with_cosine_similarity



In [ ]:

    
from IPython.core.display import HTML
css = open('table.css').read() + open('notebook.css').read()
HTML('<style>{}</style>'.format(css))



In [ ]:

    
video_games = pd.read_json("videogames.json")
video_games.head(5)



In [ ]:

    
ratings = pd.read_csv("ratings.csv")
ratings.head(5)

Filter out video games which have no title



In [ ]:

Filter out video games which have less than 5 ratings



In [ ]:

Filter out users which have less than 5 ratings



In [ ]:

Merge ratings and videogames dataframe on productID and set to df dataframe



In [ ]:

Create a pivot matrix with index userID, columns productID and values ratings



In [ ]:

Set columns to productID variable



In [ ]:

Set index to userID variable



In [ ]:



In [ ]:

    
correlated_items = ratings_pivot.corr()["B002I0JZOC"].sort_values(ascending=False).head(5)
correlated_items.index



In [ ]:



In [ ]:

    
cosine_similarity_matrix = recommend_with_cosine_similarity(ratings_pivot)["B002I0JZOC"].sort_values(ascending=False).head(5)
cosine_similarity_matrix.index



In [ ]:



In [ ]:



In [ ]:



In [ ]:



In [ ]:



In [ ]:

Filter out video games which have no title

Filter out video games which have less than 5 ratings

Filter out users which have less than 5 ratings

Merge ratings and videogames dataframe on productID and set to df dataframe

Create a pivot matrix with index userID, columns productID and values ratings

Set columns to productID variable

Set index to userID variable

Recommend video games with pearsonR

Recommend video games with cosine similarity

Check if out recommendations in related column of dataset