Yelp Analysis Project
The purpose of this project was to try to predict how many stars a user gave to a business from the comment a user wrote in their review. For example, a user writes "great restaurant thank you!" and we want to predict the user gave a 5 star rating from that comment. From the comment we used sentiment analysis, text length, business category, and geographic location of the business.Â
Click Here To Download The Dataset
The point of predicting business ratings is to understand the customer bases across different geographic regions in the United States. From a deeper understanding of how customers write reviews in relation to ratings, we can understand how to improve businesses' ratings and help them stand out against other businesses. For example, we found that in Illinois if longer reviews are given then it is likely there is a lower score while in California longer reviews are associated with a better score. This provides insights into the two different cultures between states that businesses can use to make customers better satisfied.
Rating Prediction Presentation
Data Science Final Presentation.pdfRating Prediction Final Report
Final Paper.pdf