Fusionflix is a movie recommendation system that provides movie recommendations based on the content of movies. It leverages a movie dataset, processes the data to extract meaningful tags, and uses vectorization and the bag-of-words model to generate recommendations. Additionally, Fusionflix integrates with the TMDB API to fetch movie images, and the frontend is built using Streamlit.
- Content-Based Recommendations: Uses movie metadata to provide recommendations based on content similarity.
- Data Preprocessing: Preprocesses movie data to extract tags for better recommendation accuracy.
- Vectorization and Bag-of-Words: Utilizes vectorization and the bag-of-words technique to compute movie similarities.
- TMDB API Integration: Fetches movie images using the TMDB API.
- Streamlit Frontend: Interactive user interface built with Streamlit.
-
Clone the repository:
git clone https://github.com/yourusername/Fusionflix.git cd Fusionflix
-
Install the dependencies:
pip install -r requirements.txt
-
Set up environment variables: Create a
.env
file in the root directory of the project and add your TMDB API key:API_KEY=your_tmdb_api_key
-
Run the Streamlit app:
streamlit run app.py
-
Interact with the application: Open your browser and go to
http://localhost:8501
to start using Fusionflix. Enter a movie name to get personalized recommendations and view movie images fetched from the TMDB API.
The recommendation engine works by:
- Calculating Similarities: Compute cosine similarities between movie vectors.
- Generating Recommendations: Recommend movies based on the highest similarity scores.
This project is licensed under the MIT License - see the LICENSE file for details.