ASPECT EXTRACTION & SEGMENTATION IN OPINION MINING

Abstract

Opinion mining or sentiment analysis is the computational study of people’s opinions, appraisals, attitudes, and emotions towards their aspects. Opinion mining is one of the significant areas of research in Web mining and Natural Language Processing (NLP) recently. With the growth of e-commerce, web documents are increasingly being used for decision making by individuals and organizations. This paper focuses on the identification of aspects related to customer opinions. Most of the recent works concentrate on explicit aspects only. Very few of them have dealt with implicit aspects. Here both explicit as well as implicit aspect is considered.  A multi aspect review sentence will be segmented into multiple single aspects by segmentation because different opinions can be expressed on multiple aspects simultaneously in the same review. Opinions are polled to determine positive or negative comments for efficient decision making.