Multi-aspects Rating Prediction Using Aspect Words and Sentences

In this paper we propose a method for a rating prediction task. Each review consists of several ratings for a product, namely aspects. To predict the ratings of the aspects, we utilize not only aspect words, but also aspect sentences. First, our method detects aspect sentences by using a machine learning technique. Then, it incorporates words extracted from aspect sentences with aspect word features. For estimating aspect likelihood of each word, we utilize the variance of words among aspects. Finally, it generates classifiers for each aspect by using the extracted features based on the aspect likelihood. Experimental result shows the effectiveness of features from aspect sentences.