Sentiment clustering of product object based on feature reduction

Face to car domain, the content of reviews is very scattered. In order to evaluate rate of each product, we extract the sentences from reviews. Then, those sentences are merged which describe the same product together and summarizing them according to product performances. On this basis, we extract features from these sentences, and calculate their value. Owing to existing missing feature values, established information systems are called incomplete information systems. For the problems of high dimension and missing data, we adopt the feature reduction algorithm based on discernibility matrix to reduce the feature dimension. Lastly, we aggregate each product by K-means clustering algorithm. Our experimental results indicate that the proposed method is effective.