点击次数：
论文类型：期刊论文
发表时间：2014-07-15
发表刊物：Journal of Computational Information Systems
收录刊物：EI、Scopus
卷号：10
期号：14
页面范围：5923-5930
摘要：Nowadays shopping online is very popular in the world wide. And people are apt to make decisions based on the Web information, of which word-of-mouth (WOM) rankings of products play an important role. This paper aims to explore an effective WOM ranking method by means of ratings and comparative votes on products. It firstly modifies the original ratings to generate a more reasonable ranking and then combines modified ratings and comparative votes to form the summarized scores for all products. To represent the comparative correlations between each pair of products, a weighted graph is built on which the WOM of products are quantified as scores. As a result, a ranking list based on summarized scores for the given products is gained. Evaluation on the proposed WOM ranking method is conducted by using the real world datasets coming from a famous website in China. Empirical results show that the WOM ranking list obtained by our approach has a higher support than the typical rating based method and the modified rating based method. ? 2014 by Binary Information Press