└── recommend_movies


/recommend_movies:
--------------------------------------------------------------------------------
 1 | import pandas as pd
 2 | 
 3 | # Load movie ratings data (replace with your data source)
 4 | ratings_data = pd.read_csv("movie_ratings.csv")
 5 | 
 6 | # Create a pivot table to show user-movie ratings
 7 | ratings_matrix = pd.pivot_table(ratings_data, values='rating', index='user_id', columns='movie_id')
 8 | 
 9 | # Function to recommend movies based on user's ratings history
10 | def recommend_movies(user_id, ratings_matrix, num_recommendations=5):
11 |     # Get the user's ratings
12 |     user_ratings = ratings_matrix.loc[user_id]
13 | 
14 |     # Find movies similar to those the user has rated highly
15 |     similar_movies = ratings_matrix.corrwith(user_ratings)
16 |     similar_movies = similar_movies.drop(user_id)  # Exclude user's own ratings
17 |     similar_movies = similar_movies.fillna(0)  # Fill missing values with 0
18 | 
19 |     # Sort similar movies by correlation coefficient (higher is better)
20 |     similar_movies = similar_movies.sort_values(ascending=False)
21 | 
22 |     # Recommend the top N movies
23 |     return similar_movies.head(num_recommendations).index.to_list()
24 | 
25 | # Example usage:
26 | user_id = 123  # Replace with the user's ID
27 | recommended_movies = recommend_movies(user_id, ratings_matrix)
28 | print(f"Recommended movies for user {user_id}: {recommended_movies}")
29 | 


--------------------------------------------------------------------------------