This project is a movie recommender system built using Python and Streamlit for the frontend. The system suggests movies similar to the one selected by the user, based on a precomputed similarity matrix.
https://utk-movie-recommender.streamlit.app/
- Movie Recommendations: Select a movie, and the system will recommend up to 14 similar movies.
- Streamlit Frontend: A user-friendly interface allows for easy interaction and real-time results.
- Recommendation Logic: Uses cosine similarity between movie features for recommendations.
- Python: Backend logic for the recommendation engine.
- Streamlit: Frontend for displaying movie recommendations.
- Pandas: Data manipulation and handling.
- Numpy: Used for numerical operations.
- Scikit-learn: For calculating cosine similarity.
- NLTK: Natural Language Tokeniser use for stemming.
- Pickle: For loading preprocessed data and similarity matrix.
From the dropdown list, select a movie you have watched.
Click the "Recommend" button to receive a list of similar movies based on a precomputed similarity matrix using cosine similarity.
- The system loads movie data and a similarity matrix from pickle files.
- For the selected movie, it computes movie similarities using cosine distance, and recommends the top 14 similar movies.
First, clone the repository to your local machine:
git clone https://github.com/your-repo/movie-recommender.gitNavigate to the project directory and install the required dependencies using the requirements.txt file:
cd movie-recommender
pip install -r requirements.txtTo launch the Streamlit app, run the following command in your terminal:
streamlit run app.pyThis will start a local server, and you'll see a URL where you can access the app, typically: http://localhost:8501/