Book Recommendation System Using Goodreads Dataset

Goals: Create an AI book recommendation system based on the Goodreads raw review dataset for spoiler detection.

Results: See below. Project work is still ongoing

Motivation

This project was part of a Physics 5680 (Big Data Analytics for Physics) class. The project was to create an AI book recommendation system based on the Goodreads dataset. This was a pre-defined project choice out of many from the course. Below is a pipeline of the project.

Pipeline

Report

A report was required as part of the project. The report outlines all aspects of the project. It is attached below. The report includes information like the training of BERT-Tiny, the confusion matrix, the pipeline, the collaborative filtering, and the results of the project.

Project IPYNB

Below is the PDF version of the Jupyter Notebook creating the project. It is attached below for viewing.

See the full project on Github