Journals Proceedings

Abstract

This paper presents a sentiment analysis model which combines conceptual graph enriched by domain ontology which expert-specified operational sentiment rules.Whereas conceptual graph and domain ontology allows automatic shalow parsing of natural sentences, operatioanal sentiment rules allows the sentiment system captures the linguistic knowledge provided by experts and integrates this knowledge seamlessly into the sentiment analysis process. We found that domain ontologies are extremely helpful when applied in specific domains. In the meantime, operational sentiment rules make our system achieve better performance as compared to data mining techniques.