Mining all the useful reviews from the social networks is the attracting research topic in recent years. It is necessary to analyze the reviews to know about the product or about the topic. Existing works only focus on data extraction and classifying the reviews into positive or negative and spam or not spam. The review relevance, which is very important to rank reviews, has not been considered in the existing work. In this paper, we put forward a method to calculate review relevance. We calculate the review relevance value not only by considering the similarity and correlation, but also the votes for each review. It is proved that our method works well in terms of effectiveness and accuracy. Thus, this work helps in effectively retrieving the useful reviews from social networks.
展开▼
