Identification of Features from User Opinions using Domain Relevance

Abstract

Identification of opinion features from online user reviews is a task to identify on which feature user is going to put his opinion. There are number of existing techniques for opinion feature identification but, they are extracting features from a single corpus [2]. These techniques ignore the non trivial disparities in distribution of words of opinion features across two or more corpora. This work discusses a novel method for opinion feature identification from online reviews by evaluation of frequencies in two corpora, one is domain-specific and other is domain-independent corpus. This distribution is measured by using domain relevance [12]. The first task of this work is the identify candidate features in user reviews by applying a set of syntactic rules. The second step is to measure intrinsic-domain relevance and extrinsic-domain relevance scores on the domain dependent and domain-independent corpora respectively. The third step is to extract candidate features that are less generic and more domain specific, are then conformed as opinion features. This approach is called as intrinsic extrinsic domain relevance.
DOI: 10.17762/ijritcc2321-8169.1506113