Projects of thesis codes I helped for some master students.

This repo contains all files needed for building a recommender system based on 2019 Yelp Challenge Datasets. This is the No.1 solution in USC Viterbi Data Mining Competition.

Basic movie recommender system using item based collaborative filtering

基于ItemCF与Springboot的图书商城系统

in this section will be item based recommender on movies and ratings dataset

In this section, I will create a item-based recommender on the movie dataset

Using the MovieLens 20 Million review dataset, this project aims to explore different ways to design, evaluate, and explain recommender systems algorithms. Different item-based and user-based recommender systems are showcased as well as a hybrid algorithm using a modified page-rank algorithm.

Building a collaborative filtering recommender systems on books dataset

Item-Based collaborative filtering with KNN algorithm about hotel recommendations

Recommendation System for an Online Beer Company

TMDB_5000_Movie_recommendation_system is a repository for a hybrid movie recommendation system. Discover personalized movie recommendations based on user preferences and movie features using the TMDB 5000 Movies dataset.

A dashboard to discover and search for Korean TV series. Built using React, Flask, SCSS, Sklearn and Docker.

USC DSCI 553 - Foundations & Applications of Data Mining - Spring 2024 - Prof. Wei-Min Shen

This repo has an implementation of popular recommendation techniques like user-based and item-based collaborative filtering techniques for recommending books and music.

Recommender system for board games built on data collected from major board game forum, BoardGameGeek.

Building a recommendation system and deplying using streamlit

This project aims to build a Book recommendation system using methods such as Model, Collaborative, and Content-based filtering.

Project Overview: Knowledge-based, Content-based and Collaborative Recommender systems are built on MovieLens dataset with 100,000 movie ratings.

The project's goal is to create diverse recommendation systems that predict user-item ratings