It has been showed that over last past 10 years, global retail e-commerce has been explosive growth. Without experience of product commonalities, customers will heavily rely on other reviewers for the decision making for the online shopping. However, with amount of information from reviewers, it is important for a website to identify the useful, valid, and reliable reviews to save help customers. Useful reviews also can in turn to help business to improve their products and understand customers' need.
The goal of this project is to mine 9,000,000 book reviews from Amazon.com to understand the features of reviews for helpfulness and develop preliminary classification models to predict helpfulness of a review foods