Book Recommender System

Project Overview

An end-to-end machine learning system designed to deliver personalized book recommendations by analyzing user reading patterns and book ratings using collaborative filtering algorithms.

Key Features

• Collaborative Filtering: Implemented matrix factorization and nearest neighbor algorithms to discover user-item interaction similarities.
• Interactive Web Interface: Developed an intuitive, responsive frontend using Streamlit for instant user queries and visual book discovery.
• Cloud Deployment: Containerized and deployed on AWS infrastructure ensuring high availability and low-latency inference.
• Data Pipeline: Automated data cleaning, preprocessing, and feature transformation pipelines for large-scale book metadata and user ratings.

Tech Stack

• Languages & Frameworks: Python, Streamlit, Scikit-learn, Pandas, NumPy
• Deployment & Cloud: AWS (EC2), Docker