Publication Date

Spring 2023

Degree Type

Master's Project

Degree Name

Master of Science (MS)


Computer Science

First Advisor

Katerina Potika

Second Advisor

Robert Chun

Third Advisor

William Andreopoulos


recommendation systems


Yelp is a popular social media platform that has gained much traction over the last few years. The critical feature of Yelp is it has information about any small or large-scale business, as well as reviews received from customers. The reviews have both a 1 to 5 star rating, as well as text. For a particular business, any user can view the reviews, but the stars are what most users check because it is an easy and fast way to decide. Therefore, the star rating is a good metric to measure a particular business’s value. However, there are other attributes available on the platform that can be used to enhance recommendations.

In this project, we hypothesize that by considering six different attributes of reviews, users, and businesses we can enhance recommendations. Based on these at- tributes we generate an overall popularity score for each business. Furthermore, this popularity score is the possible identifier of the business’s value. We perform experi- ments on a Yelp available dataset, by using Natural Language Processing techniques, and neural network approaches.