MULTI-STRATEGY SENTIMENT ANALYSIS OF CONSUMER REVIEWS WITH PARTIAL PHRASE MATCHING

Sentiment analysis is useful in commercial intelligence application environment and recommender systems, because it is a very convenient channel for the two ends of the supply to communicate. In the sentiment analysis, many strategies and techniques were used, such as machine learning, polarity lexicons, natural language processing, and psychometric scales, which determine different types of sentiment analysis, such as assumptions made, method reveals, and validation dataset. Since Internet has become an excellent source of consumer reviews, the area of sentiment analysis (also called sentiment extraction, opinion mining, opinion extraction, and sentiment mining) has seen a large increase in academic interest over the last few years. Sentiment analysis mines opinions at word, sentence, and document levels, and gives sentiment polarities and strengths of articles. As known, the opinions of consumers are expressed in sentiment phrases. Traditional machine learning techniques can not represent the opinion of articles very well. This project proposes a multi-strategy sentiment analysis method with semantic similarity to solve the problem with partial phrase matching. Naïve Bayes classification is also applied to find the probability of data distribution in various category of data set. The project is designed using R Studio 1.0. The coding language used is R 3.4.4.