Movie Recommendation System

• Tech Stack: Python, Pandas, NumPy, Scikit-learn, Surprise, NLTK, Matplotlib
• Project Focus: Building Multiple Movie Recommendation Engines
• Github URL: Project Link

This project demonstrates the creation of multiple recommendation engines for suggesting movies to users. The system leverages a combination of algorithms, including Simple Recommender, Content-Based Filtering, Collaborative Filtering, and a Hybrid Engine. Techniques like TF-IDF, Cosine Similarity, and Singular Value Decomposition (SVD) are used to provide highly personalized movie recommendations based on user preferences and movie features.

The project includes:

• Simple Recommender: Ranks movies based on popularity and vote averages using the IMDb weighted rating system.
• Content-Based Recommender: Provides recommendations based on movie overviews, taglines, and metadata (genre, cast, crew, keywords).
• Collaborative Filtering: Predicts user ratings for movies using the Surprise library.
• Hybrid Engine: Combines content-based and collaborative filtering for more accurate recommendations.