Book Recomendation
๐ Book Recommendation System
๐ Project Description
This is a Flask-based web application that recommends books based on popularity and ratings. The system is built using a collaborative filtering approach and primarily utilizes a popularity-based model to suggest books with the highest ratings.
๐ Live Demo โ Book Recommendation System
๐ GitHub Repository
๐ How to Use
1๏ธโฃ Visit the homepage to see the top recommended books.
2๏ธโฃ Enter the book name or search criteria in the input box.
3๏ธโฃ Click โRecommendโ to see the suggestions.
๐ Dataset
The project uses a book recommendation dataset from Kaggle for training and evaluation.
๐ Kaggle Dataset Link โ (Provide actual link here)
๐ Model Training Approach
The book recommendation system is trained using the following steps:
1๏ธโฃ [popular_book] โ Identifies the top 50 most popular books based on ratings and votes.
2๏ธโฃ [stan_df] โ Filters users who have rated more than 200 different books.
3๏ธโฃ [rating_50_book] โ Selects books that have received more than 50 ratings (i.e., rated by at least 50 different users).
4๏ธโฃ [pt] โ Creates a pivot table for rating_50_book
based on ratings and votes.
5๏ธโฃ [similarity_s] โ Computes the similarity score between books.
6๏ธโฃ Recommendation Function โ Given a book name, the system suggests 6 most related books.
๐ Deployment (Render)
The model is deployed online using Render.com for easy accessibility.
๐ง Deployment Steps:
1๏ธโฃ Render Setup โ Create a Render.com account and set up a new โWeb Serviceโ.
2๏ธโฃ Connect GitHub โ Link the Render service to this GitHub repository.
3๏ธโฃ Configure Build Commands โ Review and adjust the build commands (auto-detected for Python projects).
4๏ธโฃ Deploy โ Click โDeployโ, and Render will build & launch the app.
5๏ธโฃ Access โ Use the Render URL to access the deployed system.
๐ฎ Future Enhancements
๐ User-Based Filtering โ Recommend books based on individual user preferences.
๐ Content-Based Filtering โ Suggest books using textual similarities in descriptions.
๐ Hybrid Model โ Combine popularity-based & collaborative filtering for improved accuracy.
๐ Enhanced UI โ Develop a more interactive & visually appealing interface.
๐ This project helps book lovers discover new recommendations effortlessly! ๐