tonybhaskar / Recommendation-System

Movie Recommendation System

This is a Movie Recommendation System built using Streamlit, which recommends movies based on the similarity to a selected movie. The system provides the top 5 recommendations along with their posters.

Features

• Movie Selection: Select a movie from a dropdown list.
• Recommendations: Get the top 5 recommended movies based on similarity.
• Poster Display: View the posters of the recommended movies.

Installation

1. Clone the Repository:

   git clone https://github.com/TonyBhaskar/Recommendation-System.git
   cd Recommendation-system

2. Install Dependencies: Ensure you have Python installed, then install the required packages:

   pip install streamlit pandas requests

3. Download Required Files:

   • Ensure you have the movies.pkl and similarity.pkl.gz files in the root directory.

Usage

1. Run the Application:

   streamlit run app.py

2. Select a Movie:

   • Use the dropdown to select a movie.

3. Get Recommendations:

   • Click the "Recommend" button to see the top 5 movie recommendations along with their posters.

File Structure

• app.py: Main application script for the Streamlit app.
• movies.pkl: Pickle file containing the list of movies.
• similarity.pkl.gz: Gzipped pickle file containing the similarity matrix.
• Main.ipynb: Jupyter notebook file, presumably used for data preprocessing or model building.

Dependencies

• streamlit: For building the interactive web application.
• pandas: For handling movie data.
• requests: For fetching movie posters from the TMDB API.
• gzip: For decompressing the similarity matrix file.
• pickle: For loading serialized data.

API

• The Movie Database (TMDB): Used to fetch movie posters.

Contributing

Contributions are welcome! Feel free to open an issue or submit a pull request.

License

This project is licensed under the MIT License - see the LICENSE file for details.