Identification of Opinionated Features Extraction from Unstructured Textual Reviews
Haritha Donavalli1, Balaji Penubaka2
1Haritha Donavalli, Professor & Head BES, Department of CSE, KL University, Guntur (Andhra Pradesh), India.
2Balaji Penubaka, Research Scholar, Department of CSE, KL University, Guntur (Andhra Pradesh), India.
Manuscript received on 23 April 2019 | Revised Manuscript received on 05 May 2019 | Manuscript Published on 17 May 2019 | PP: 136-139 | Volume-7 Issue-6S4 April 2019 | Retrieval Number: F10260476S419/2019©BEIESP
Open Access | Editorial and Publishing Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Now a days online marketing rapidly growing day by day. Most of the customers are very interested to study the product reviews before buying any product through online shopping. In this regard, opinion mining or sentiment analysis plays the major roles to extract various product attributes to give rank the sellers as well as products. In this paper, we recommend a innovative technique to classify the opinion features through reviews by using domain specific collections and domain independent collections respectively. Domain relevance (DR) is the difference between domain dependent collection and domain independent collection, and is relevant term or word in the document reviews. We primarily use the syntactic dependency principles for candidate features extraction. We can compute Intrinsic domain relevance (IDR) evaluation score based on domain dependent collection and extrinsic domain relevance (EDR) evaluation score based on domain independent collection respectively. Main features of review document got lesser conventional EDR score and greater IDR score more than other cut-off finalized as opinionated features.
Keywords: Opinion Mining, Domain Relevance, Sentiment Analysis, Intrinsic & Extrinsic Domain Relevance.
Scope of the Article: Software Domain Modelling and Analysis