A comprehensive study on opinion mining features and their applications

Abstract

E-shopping is a modern approach to buy or sell products/services and becomes so popular with growing the Internet. So, opinion mining becomes a very important concept in data mining area as researchers and business usually need to know about the overall sentiment of viewpoint of people about desired phenomena. Opinion mining is the fundamental phase for variety data mining applications such as opinion summarization, recommendation system and opinion spam detection. To achieve the best results of opinion mining, we need to use the proper set of features for classification and clustering. In this paper, we do comprehensive investigation on various types of features exploited in variety sub-branches of opinion mining domain. We present the most frequent features sets includes structural, linguistic and relation-based features as a comprehensive reference for further opinion mining research. The results proved that using multiple types of features improve the accuracy of opinion mining applications.