Weighing Stars: Aggregating Online Product Reviews for Intelligent E-commerce Applications

Abstract

A new task in text-sentiment analysis adds usefulness scoring to polarity/ opinion extraction to improve product- review ranking services, helping shoppers and vendors leverage information from multiple sources. Human language is a medium not only for exchanging information but also for conveying subjective opinions and emotion. Recently, interest in text-subjectivity and sentiment analysis has increased as part of the larger research effort in affective computing, which aims to make computers understand and generate human-like emotions through language and other expressive activities such as gesture.