TV Show Recommendation Application

This code provides a recommendation system based on content-based filtering using the user-item matrix. It predicts ratings for anime shows and generates personalized recommendations for users.

Prerequisites

• Python 3.x
• pandas
• scikit-learn
• NumPy
• FuzzyWuzzy
• Streamlit

Data

The code assumes the presence of the following CSV file in the current working directory:

• anime.csv: Contains information about anime shows, including anime_id, name, genre, type, episodes, rating, and members.

Instructions for Use

1. Ensure that the required CSV file anime.csv are present in the same directory as the Python script
2. Run the code ensuring that 2 pickle files (animes_list.pkl and similarity.pkl) are generated
3. In the terminal, enter the command streamlit run app.py
4. Enter the desired anime title into the search box located in the application window to receive the top 5 personalized recommendations for similar anime titles.
   • Click on the poster titles to be redirected to a page for more information on that title

Demo

demo.mp4