Title: Mining explicit and implicit opinions from reviews

Authors: Farek Lazhar; Tlili-Guiassa Yamina

Addresses: Computer Science Department, Badji-Mokhtar Annaba University, BP 12, 23000, Annaba, Algeria ' Computer Science Department, Badji-Mokhtar Annaba University, BP 12, 23000, Annaba, Algeria

Abstract: The huge amount of subjective data available on the web carrying people's opinions, sentiments and beliefs, is an important resource for companies and merchants who want to ameliorate their products and services and for individuals who are interested in other's opinions for purchasing a product or using a service, finding opinions on political topics, etc. This paper presents an approach for opinion extraction. It mines reviews to extract features-opinion pairs and then classify the opinionated features into one of two main classes: positive or negative. Our approach is articulated on the use of dependency grammar to extract explicit feature-opinion pairs and the use of domain ontology to extract implicit feature-opinion pairs by exploiting relations between concepts, individuals and attributes. Finally, the classification task is guided by support vector machine (SVM) as a supervised learning technique.

Keywords: opinion mining; extraction; ontology; dependency grammars; feature extraction; classification; data mining; explicit opinions; implicit opinions; online reviews; support vector machines; SVM; supervised learning.

DOI: 10.1504/IJDMMM.2016.075966

International Journal of Data Mining, Modelling and Management, 2016 Vol.8 No.1, pp.75 - 92

Available online: 20 Apr 2016 *

Full-text access for editors Access for subscribers Purchase this article Comment on this article