Abstract

This paper presents a feature-driven opinion summarization method for customer reviews on the web based on identifying general features (characteristics) describing any product, product specific features and feature attributes (adjectives grading the characteristics). Feature attributes are assigned a polarity using on the one hand a previously annotated corpus and on the other hand by applying Support Vector Machines Sequential Minimal Optimization[1] machine learning with the Normalized Google Distance[2]. Reviews are statistically summarized around product features using the polarity of the feature attributes they are described by.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call