first_time_recommender_system.py

# -*- coding: utf-8 -*-
"""
Created on Fri Feb  8 23:49:57 2019

@author: AD
"""

import numpy as np 
import pandas as pd 
import matplotlib.pyplot as plt
from sklearn.metrics import pairwise_distances
from scipy.spatial.distance import cosine, correlation


u_cols = ['user_id', 'age', 'sex', 'occupation', 'zip_code']
users = pd.read_csv('u.user', sep='|', names=u_cols,
                    encoding='latin-1', parse_dates=True) 

r_cols = ['user_id', 'movie_id', 'rating', 'unix_timestamp']
ratings = pd.read_csv('u.data', sep='\t', names=r_cols,
                      encoding='latin-1')

m_cols = ['movie_id', 'title', 'release_date', 'video_release_date', 'imdb_url']
movies = pd.read_csv('u.item', sep='|', names=m_cols, usecols=range(5),
                     encoding='latin-1')

movie_ratings = pd.merge(movies, ratings)
df = pd.merge(movie_ratings, users)

df.head(2)




ratings.head(1)
movies.head(1)
users.head(1)



df.drop(df.columns[[3,4,7]], axis=1, inplace=True)
ratings.drop( "unix_timestamp", inplace = True, axis = 1 ) 
movies.drop(movies.columns[[3,4]], inplace = True, axis = 1 )
#Dropping all the columns that are not really needed
df.info()




movie_stats = df.groupby('title').agg({'rating': [np.size, np.mean]})
movie_stats.head()


min_50 = movie_stats['rating']['size'] >= 50
movie_stats[min_50].sort_values([('rating', 'mean')], ascending=False).head()



ratings.rating.plot.hist(bins=50)
plt.title("Distribution of Users' Ratings")
plt.ylabel('Number of Ratings')
plt.xlabel('Rating (Out of 5)');

users.age.plot.hist(bins=25)
plt.title("Distribution of Users' Ages")
plt.ylabel('Number of Users')
plt.xlabel('Age');

          
ratings_matrix = ratings.pivot_table(index=['movie_id'],columns=['user_id'],values='rating').reset_index(drop=True)
ratings_matrix.fillna( 0, inplace = True )
ratings_matrix.head()