IDENTIFYING FEATURES IN OPINION MINING FOR SENTIMENT SUMMARIZATION

COMPUTER SCIENCE

Abstract— The E-commerce application contains opinion mining are the process of brings out the emotions of the public. The customer opinions are composed through the Online Shopping websites such as Amazon, Flip kart etc., and the opinion mining are positive or negative using Word-Alignment Model. Now-a-days the most emphasized subjects are under opinion mining and some of the controlling this approaches are duplicate comments. The recommendations and opinion in review sites are used for marketing and give the mindfulness to individual people. The peoples can feel free to share their attitudes, ideas, suggestion whether it may be positive or negative. The purpose of the work is to increase the accuracy of the result in manufactured goods review through this give quality of product to millions of peoples and also predict the online customer preference and also gives the survey rating to the product. This paper mainly focuses on review sites and analyzes the opinion target and opinion word extractions are not new tasks in opinion mining based on Clustering based on Frequent Word Sequences (CFWS) algorithm. There is an important determination absorbed on these tasks. They can be separated into two categories: sentence-level extraction and corpus level extraction according to their extraction aims. The experimental results show that this approach method improves performance over the traditional methods.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.