content-based-recommender-system

Lightweight content-based movie recommender that converts movie metadata (titles, genres, descriptions) into TF‑IDF vectors and ranks films by cosine similarity. Clean, modular Python code with example notebooks and minimal dependencies—easy to extend and integrate.

Hybrid movie recommendation system combining Collaborative Filtering (ALS), Content-Based Filtering (TF-IDF & Sentence-Transformers + TMDB metadata), and MMR diversity reranking — with a rigorous offline evaluation harness benchmarking 5 models on MovieLens 1M.

A content-based recommender system designed to help Communication Science Master's students at the UvA find suitable thesis supervisors based on their research interests.

By 2020, Board Game Geek featured over 36,000 board games, with 5,000 new games introduced that year alone. To manage this rapid influx and effectively connect players with the right games, the team developed a Meta-level hybrid recommendation system for the Board Game Geek platform.

Recommend movies based on user preferences

Hybrid movie recommendation system combining Prolog-based knowledge representation, content similarity and user ratings.

A content-based Movie Recommender System built using Machine Learning that recommends similar movies based on user selection.The application uses cosine similarity to generate recommendations and displays movie posters through the TMDB API .It features an interactive and user0friendly interface developed with Streamlit

A content-based recommender system using TF-IDF and cosine similarity to suggest personalized items based on user profiles.

A movie manager web app, with CLI support

This project delivers a seamless recommendation experience by blending machine learning similarity models with The Movie Database (TMDB) API for real-time posters, summaries, and release details. Lightweight, fast, and designed for production-grade deployment.