Multi-aspects review summarization with objective information
Kazutaka Shimada, Ryosuke Tadano, Tsutomu Endo · Procedia - Social and Behavioral Sciences · 2011
In this paper, we propose a method for multi-aspects review summarization based on evaluative sentence extraction. We handle three features; ratings of aspects, the tfidf value, and the number of mentions with a similar topic. For estimating the number of mentions, we apply a clustering algorithm. By using these features, we generate a more appropriate summary. In this paper, we also focus on objective information of the target product. We integrate the summary from sentiment information in reviews and the objective information extracted from Wikipedia. The experiment results show the effectiveness of our method.