/Collaborative-Filtering

Primary LanguageJupyter Notebook

Movie Recommendation System

This project implements a collaborative filtering recommendation system to predict movie ratings and recommend movies to users.

Dataset

The dataset consists of two files:

  • ratings.csv: Contains user ratings for movies.
  • movies.csv: Contains metadata about movies.

Methodology

  1. Data Preparation: Merge the ratings and movies datasets.
  2. Normalize Ratings: Normalize the user-item matrix.
  3. Calculate Similarity: Compute item similarity using Pearson correlation.
  4. Make Predictions: Predict ratings for movies that a user has not rated.