A unified machine learning application that provides personalized recommendations across three domains: Books, Movies, and Music. Built using Python, ML techniques, and deployed with Streamlit for an interactive user experience.
-
Books Recommender
- Suggests books based on title and author similarity using TF-IDF vectorization and cosine similarity.
-
Movies Recommender
- Recommends movies using precomputed content similarity over genres, keywords, and more.
-
Music Recommender
- Finds similar Spotify tracks based on song name, artist, and genre.
- Optimized to handle large datasets (600K+ songs) with efficient TF-IDF + cosine similarity on demand.
-
Streamlit Web Interface
- Home page to select between books, movies, and music.
- Dedicated pages showing recommendations with similarity scores and graphical insights.
- Python 3.11
- Pandas, NumPy for data handling
- Scikit-learn for TF-IDF & cosine similarity
- Matplotlib for plots
- Streamlit for interactive UI
Access the website : https://kkavy-project.streamlit.app/