Development of a book recommender system

Ling, Shi Shi (2025) Development of a book recommender system. Final Year Project, UTAR.

PDF Download (5Mb)

Abstract

This project aims to create a hybrid book recommendation system that addresses key challenges such as usability and customization. Existing applications frequently fall short of providing tailored and intuitive experiences, leaving a gap for customers who want more control over their preferences. By combining collaborative filtering, which leverages user behavior like reading history and ratings, with content-based filtering that analyzes book metadata such as genre, author, and title, the system solves limitations such as data sparsity and the cold start problem. The report's literature review part includes reviews of applications such as Goodreads, Meet New Books, and StoryGraph. A review of existing applications demonstrates the need for a more flexible and user-centered approach. The proposed solution is a mobile application built with Flutter, Python, Firebase, and PostgreSQL that dynamically adjusts recommendations depending on ongoing user feedback. The system uses weighted hybrid methodologies to provide accurate, diverse and personalized book suggestions, with the goal of increasing user satisfaction and transforming the book discovery experience.

Actions (login required)