科研之友
微信
新浪微博
Facebook
分享链接
摘要
There is a growing interest in sharing personal opinions on the Web, such as product reviews, economic analysis, and political polls. Existing research focuses on document-based approaches and documents are represented by the bag-of-words model. However, due to the loss of contextual information, this representation fails to capture the associative information between an opinion and its corresponding target. Additionally, several researchers focus on sentence-based approaches, which can effectively extract an aspect-sentiment word pair within one sentence. Nevertheless, their approaches can only deal with one aspect within one sentence and miss the identification of sentiment modifier. This paper firstly proposes a novel identification approach of aspect-modifier-sentiment word triple using shallow semantic information. Our approach can effectively deal with multi-aspect in one sentiment sentence. Secondly, for the quantitative analysis of aspect-modifier-sentiment word triple, we present an improved sentiment word quantitative method to generate sentiment score for every word in sentiment lexicon. Experimental results show that our approach is feasible and effective.
