Recommender Systems for Deal Websites
Our Team
Principle Investigators
- PI: Lucy Kerns (webpage), Mathematics and Statistics
- Co-PI: Dr. Feng (George) Yu (webpage), Computer Science and Information Systems
Undergraduate Researchers
Project Description
A recommender system is a machine learning technique that is in wide use by almost every big company, especially in the e-commerce area. In general, a recommender system serves the purposes of presenting a ranked list of objects given input object and providing customers with information to help them find relevant items.
The focus of the proposed project is to build and evaluate different recommendation algorithms - including the traditional collaborative recommender systems based on association rules and more recent collaborative algorithms based on dimensionality reduction - that are particularly suited for datasets such as those that are commonly collected by deal websites. To do so, we will use two datasets. The first data set will be collected from deal websites, and the second data set will be collected through a survey system proposed by our research team.
The primary goal of this proposed project is to engage undergraduate students in research involving recommender systems, particularly the collaborative filtering recommender systems, and to encourage them to develop independent critical thinking and problem-solving skills that are essential for further study or professional work.
Acknowledgment
This project is partially supported by Collaborative Research Experiences for Undergraduates (CREU) Grant.