Nithin, Y R and Poornalatha, G (2017) Feature Based Opinion Mining For Restaurant Reviews. Advances in Intelligent Systems and Computing, 678 (1). pp. 305-318. ISSN 2194-5357

PDF 3223.pdf - Published Version Restricted to Registered users only Download (3MB) | Request a copy

Abstract

Product reviews or customer feedback has become a platform for retailers to plan marketing strategy and also for new customers to select their appropriate product. Since the trend of e-commerce is increasing, an amount of customer reviews also has been increased to a greater extent. Consequently, it becomes a tough task for retailers as well as customers to read the reviews associated with the product. Sentiment analysis resolves this issue by scanning through free text reviews and providing the opinion summary. However, it does not provide detailed information, such as features on which the product is reviewed. Feature-based sentiment analysis methods increases the granularity of sentiment analysis by analyzing polarity associated with features in the given free text. The main objective of this work is to design a system that predicts polarity at aspect level and to design a score calculating scheme that defines the extent of polarity. Obtained feature - level scores are summarized according to users’ priority of interest

Item Type:	Article
Uncontrolled Keywords:	Natural Language Processing, aspects, reviews, free text, star rating
Subjects:	Engineering > MIT Manipal > Information and Communication Technology
Depositing User:	MIT Library
Date Deposited:	30 Sep 2017 05:01
Last Modified:	30 Sep 2017 05:01
URI:	http://eprints.manipal.edu/id/eprint/149754

Actions (login required)

View Item