Student: Kenny Kwock
Project: Fair Two-sided Spam Detection Using Machine Learning
View: Research Poster (PDF)
Institution: Lehigh University
Major: Computer Science and Engineering
Advisor: Sihong Xie
Abstract
It is not surprising to expect anyone to check online reviews before making a purchase. However, with the increasing reliance of online reviews to make a decision, many businesses seek to either promote their own business, or destroy competing businesses by employing fake review bots. Additionally, studies have shown that larger businesses receive more leniency than smaller businesses, based on their review count. With online product review aggregating websites such as Amazon and Yelp optimizing their search engines to recommend businesses based on their positivity and popularity rate, there is an increasing concern for illegitimacy and small businesses who struggle to profit. Utilizing graphical neural networks and multi-objective optimization, this paper will discuss a Machine Learning approach that seeks to optimize two sides of this issue: fake review spam detection, and business fairness.
About Kenny Kwock
Kenny Kwock is a Lehigh junior student, enrolled in a dual-degree in Computer Science & Mathematics, with interests in Probability, and Machine Learning. Last summer, he did an REU with Professor Sihong Xie, and continues his research project on Fair Spam Detection with Machine Learning. This upcoming summer, he will be doing research at a National Laboratory, and hopes to further pursue his research interest in Machine Learning in graduate school. Beyond academics, he works at Wilbur Powerhouse, the Additive Manufacturing lab, assisting students with 3D printing and designing. In his free time, he likes to design freely at the AM lab, and play Virtual Reality.