Bookster

Your Smart Book Recommendation System

---

Overview

Bookster is a personalized book recommendation system that leverages hybrid recommendation techniques, combining collaborative and content-based filtering to deliver tailored book suggestions. This project was developed as an academic exercise for a school course in Data Mining and Machine Learning. While inactive now, the project showcases the application of advanced algorithms in recommendation systems, alongside clean and responsive front-end design.

---

Features

• Personalized Recommendations: Provides book suggestions based on user preferences, ratings, and similar user behavior.
• Hybrid Recommendation System:
  • Collaborative Filtering: Suggests books based on ratings from similar users.
  • Content-Based Filtering: Recommends books based on genre, author, and metadata.
• Interactive User Interface: Built using Angular for a Netflix-style user experience.
• Backend and Database:
  • Flask backend with REST APIs to serve recommendations.
  • PostgreSQL for efficient data storage and retrieval.
• Machine Learning Models: Implemented using Python libraries like Scikit-Learn, NumPy, Pandas, and SciPy.

---

Tech Stack

• Frontend: Angular, TypeScript
• Backend: Flask, Python
• Database: PostgreSQL
• Machine Learning Libraries: Scikit-Learn, NumPy, Pandas, SciPy
• Design: Figma for UI/UX prototyping

---

How It Works

1. User Data: Users provide book ratings or interact with the system.
2. Hybrid Model:
   • Collaborative filtering uses similarity metrics to find users with similar tastes.
   • Content-based filtering analyzes book metadata to match preferences.
3. Recommendation Engine: Combines results from both methods to present personalized recommendations.
4. Frontend Display: A dynamic, visually appealing interface showcases recommended books and allows users to explore further.

---

Screenshots

Landing Page	User Book Selection
Log In	Sign Up
Preferred Authors	Preferred Publishers
Home Page

---

Academic Context

This project was developed in December 2023 as part of an academic curriculum. Although inactive now, Bookster highlights the potential of machine learning in recommendation systems, demonstrating practical use cases for advanced algorithms in real-world scenarios.

---

Disclaimer

This project is no longer actively maintained and was developed for educational purposes. It serves as a portfolio piece for showcasing technical skills in machine learning, web development, and database integration.

---

Key Takeaways

• Designed and implemented a hybrid recommendation system, evaluated for accuracy and efficiency.
• Improved the recommendation engine using advanced algorithms and optimized backend queries.
• Developed a full-stack application integrating data science and web development.
• Demonstrated the ability to create scalable and visually engaging solutions.

---

This project is inactive and was completed as part of a school assignment. For more details, feel free to reach out!