Sentiment Classification Based on Ontology and SVM Classifier

The combination of using Natural Language Processing techniques (NLP), ontology based on Formal Concept Analysis (FCA) design, and Support Vector Machine (SVM) for classifying the software reviews are positive, negative or neutral.

Abstract

There are a lot of text documents on the Web which contain opinions or sentiments about an object such as software reviews, product reviews, movies reviews, music reviews, and book reviews etc. Opinion mining or sentiment classification aim to extract the features on which the reviewers express their opinions and determine they are positive or negative. In this paper we proposed an ontology based combination approach to enhance the existing approaches of the sentiment classification. We also used the supervised learning techniques for classification of the sentiments in the software reviews. This paper proposed the combination of using Natural Language Processing techniques (NLP), ontology based on Formal Concept Analysis (FCA) design, and Support Vector Machine (SVM) for classifying the software reviews are positive, negative or neutral.