🎬 Movie Recommender System (End-to-End NLP App)

📌 Project Overview

This is a content-based recommendation engine that suggests movies based on a user's preferences. By analyzing metadata—genres, keywords, cast, and crew—the system calculates similarity scores to deliver personalized recommendations.

🚀 Live Demo

Click here to view the live app!

🛠️ Tech Stack

• Machine Learning: NLP, Cosine Similarity, CountVectorizer
• Frontend: Streamlit
• API Integration: TMDB API (The Movie Database)
• Data Engineering: Pandas, Pickle

🔍 How it Works

1. Vectorization: Movie tags are converted into 5,000-dimensional vectors using CountVectorizer.
2. Similarity Engine: The app uses Cosine Similarity to find the shortest "distance" between movie vectors.
3. Real-Time API: When a movie is suggested, the app calls the TMDB API to fetch high-resolution posters for a Netflix-style UI.

📊 Deployment

The application is deployed on Streamlit Community Cloud, connected directly to this GitHub repository for continuous integration.