Book Recommendation System using Collaborative Filtering

This is a book recommendation system built using collaborative filtering techniques. The system recommends books based on user ratings and the similarity between either books (item-item filtering) or users (user-user filtering).

Table of Contents

• About
• Features
• Technologies
• Data Sources
• Installation
• Usage
• Files and Folders
• How It Works
• Future Enhancements
• Contributing
• License

About

This project implements a book recommendation system that provides personalized recommendations to users based on their historical ratings. Collaborative filtering, a popular recommendation algorithm, is used to make predictions by analyzing user-item interactions.

Features

• Personalized recommendations based on collaborative filtering (User-User & Item-Item)
• Popular books display on the homepage
• Ability to search for recommendations by entering a book name
• Flask web application with a user-friendly interface
• Preprocessed data and fast loading using .pkl files for efficient performance

Technologies

This project uses the following tools and technologies:

• Python: Core programming language
• Flask: For building the web application
• Pandas: For data manipulation
• Numpy: For numerical operations
• Pickle: For saving and loading preprocessed data
• Jinja2: For HTML templating
• Bootstrap: For styling the front-end pages

Data Sources

• Books.csv: Contains book metadata such as book ID, title, author, etc.
• Users.csv: Contains information about users.
• Ratings.csv: Contains user ratings for different books.

These CSV files are processed and saved into .pkl files for efficient data access within the web application.

Installation

1. Clone the repository

git clone https://github.com/HimanshiVerma05/BookRecommenderSystem.git
cd book-recommender-system

2. Create and activate a virtual environment

python -m venv venv
# For Windows
venv\Scripts\activate
# For macOS/Linux
source venv/bin/activate

3. Install required dependencies

Install the dependencies listed in the requirements.txt file:

pip install -r requirements.txt

4. Set up the environment

If you're running the Flask app locally, set the Flask environment variable:

# For Windows
set FLASK_APP=app.py
# For macOS/Linux
export FLASK_APP=app.py

5. Start the Flask application

Run the Flask application locally:

flask run

6. Access the app

Open your browser and go to http://127.0.0.1:5000/ to interact with the recommendation system.

Usage

1. Home Page: The homepage displays the top 50 most popular books, which are precomputed and loaded from popular.pkl.
2. Recommendation Page: Navigate to the "Recommend" page, input the name of a book, and receive personalized recommendations.
3. Backend: The backend loads preprocessed data (books, ratings, similarity scores) from .pkl files to generate recommendations.

Files and Folders

• app.py: Main Python file that defines the Flask routes and loads data from .pkl files.
• templates/: Contains the HTML templates (index.html and recommend.html) for rendering the UI using Jinja2.
  • index.html: Displays the top 50 books.
  • recommend.html: Allows users to search for book recommendations.
• static/: Folder for static assets like CSS, images, etc.
• Books.csv, Ratings.csv, Users.csv: Raw data files used for building the recommendation system.
• books.pkl, similarity_scores.pkl, pt.pkl: Preprocessed data files for faster loading and recommendation generation.
• requirements.txt: Contains the list of dependencies required to run the project.

How It Works

1. Data Preprocessing: The book, user, and rating data are loaded from CSV files and cleaned. Missing values are handled, and similarity scores are computed based on user-item interactions.

2. Collaborative Filtering:

   • User-User Collaborative Filtering: Recommends books to a user based on the similarity to other users' preferences.
   • Item-Item Collaborative Filtering: Recommends similar books based on the similarity between books (e.g., books rated highly by the same users).

3. Pickle Serialization: The preprocessed data (books, ratings, and similarity scores) are saved in .pkl files to avoid reprocessing the data each time the app is run.

4. Flask Web Application: The Flask app serves two primary pages: the homepage showing popular books and the recommendation page, where users can input a book and get recommendations.

Future Enhancements

• Add support for content-based filtering.
• Integrate a user login system to track individual users' reading histories.
• Improve the recommendation algorithm with hybrid filtering (combining collaborative filtering with content-based techniques).
• Add more dynamic user interaction features, such as rating or reviewing books directly in the app.

References

• Special thanks to Campusx tutorial - https://www.youtube.com/watch?v=1YoD0fg3_EM
• Dataset from - https://www.kaggle.com/datasets/arashnic/book-recommendation-dataset