Review Paper on Identifying Features in Opinion Mining via Intrinsic and Extrinsic Domain Relevance

2016 International Journal of Science and Research (IJSR)  
In this paper, we have planned to propose a complete unique technique to spot opinion features from on-line reviews by exploiting the two distinct opinion feature statistics across two corpora, one domain-specific corpus (i.e., the given review corpus) and one domain-independent corpus (i.e., the contrastive corpus).We capture this inequality via domain connection (DR) that characterizes the connection of a term to a text assortment. Initial list extraction of candidate opinion options is done
more » ... rom domain review corpus by following the grammar dependence rules. For every extracted candidate feature, we can estimate its intrinsic-domain connection (IDR) and extrinsic-domain connection (EDR) scores on the domain-dependent and domain-independent corpora, severally. The aim of document-level (sentence-level) opinion mining is to classify the general judgment or sentiment expressed in a personal review document. Thus, on the basis of candidate feature, the interval threshold can be used for intrinsic and extrinsic domain connection criterion. Evaluations conducted on two real-world review domains demonstrate the effectiveness of our projected IEDR approach in distinguishing opinion options.
doi:10.21275/v5i1.30121501 fatcat:u4njbkwtpnhgjdr2ilv2lg3kku