Movie Recommendation System using Collaborative Filtering and Content-Based Filtering

Overview

This project aims to develop a movie recommendation system using the MovieLens dataset. The system is designed to provide movie recommendations that users will enjoy, keeping them engaged with the platform. The project involves two main tasks:

Collaborative Filtering (Item-Based with Cosine Similarity):

Recommend movies based on users’ rating history without using any pre-built libraries.

Content-Based Filtering:

Utilize film metadata such as genre, year, and director to recommend movies based on users' preference history. Dataset

The MovieLens dataset includes:

This data set consists of:

• 100,000 ratings (1-5) from 943 users on 1682 movies.
• Each user has rated at least 20 movies.
  • Simple demographic info for the users (age, gender, occupation, zip)