Extracting Customer Reviews from Online Shopping and Its Perspective on Product Design

Abstract

This paper presents a study on how we can extract helpful review from customers and its effect on the early phase when designing a product. We present a framework for analyzing the reviews from online shopping sites by detecting helpful reviews, aspects, and top reviews. We conduct a through analysis on review comments of the Amazon site and form a novel framework which can automatically extract useful information from review documents and it can also collaboratively work between designers and opinion customers. Experimental results on helpful review identification and sentiment classification showed that the proposed model achieved promising results. We also conduct an interview with designers to assert whether or not the proposed framework is effective. The results showed that the proposed framework is helpful for designers.

Keywords

• Helpful reviews
• sentiment classification
• product design
• user requirements
• aspect sentiments
• aspect extraction