Abstracts Category : Other

Text mining on Amazon reviews to extract feature based feedback

by Sinduja Balasubramanian

Analyzing the feedback on a product or a service helps to improve the quality ofthe product or service. That way, reviews from online shopping sites (such as Amazon)not only help a consumer to buy a product but also can help a manufacturer or seller toknow the pros and cons of their product. Amazon star ratings alone is not enough forthis. One should go through the text reviews to know specifically which feature of theproduct is lacking customer satisfaction.But a product may have thousands of reviews and it???s hard for a person to gothrough all the reviews. Hence, we need a system which can give a statistical report onthe number of reviewers not satisfied with a specific feature of a product. This projectenables the user to view feature based review for a selected category of Amazonproducts. For a particular product the percentage of dissatisfied reviewers for eachmajor feature of the product can be viewed.The dataset which includes product details and customer reviews for eachproduct are collected from Amazon.com. The implementation of this system is achievedby using MongoDB and R. The statistical results that are generated by the system arevisualized with the help of Tableau software. The Amazon reviews undergo NaturalLanguage Processing and text mining to identify major features of the product. Then a videep sentiment analysis is made to identify the polarity (positive or negative) of eachreview.This project can be further developed to a user interactive web or mobileapplication where user can choose categories and products and have better visualizationof results. Implementation required integrating both data mining and artificialintelligence techniques.Advisors/Committee Members: Gordon, V. Scott.