Sentiment analysis using rule-based and case-based reasoning

Abstract

Sentiment analysis becomes increasingly popular with the rapid growth of various reviews, survey responses, tweets or posts available from social media like Facebook or Twitter. Sentiment analysis can be turned into the question of whether a piece of text is expressing positive, negative or neutral sentiment towards the discussed topic and can be thus understood as a knowledge-based classification problem. A variety of knowledge-based techniques can be used to solve this problem. The paper focuses on two complementary approaches that originate in the area of AI (artificial intelligence), rule-based reasoning and case-based reasoning. We describe basic principles of both approaches, their strengths and limitations and, based on a review of literature, show how these approaches can be used for sentiment analysis.