随着电子商务的迅速发展，人们越来越亲睐于网上购物。在网上购物之前，消费者往往会参考该产品相关的评价以决定是否购买。因此虚假评论者的识别具有非常重要的意义。基于虚假评论者和真实评论者在情感极性上存在的差异，在特征建模过程中增加了评论文本的情感特征，并结合用户之间对于特定商品之间的关系，创建了一个多边图的模型并提出了一种识别虚假评论者的方法。实验结果验证了该算法的有效性。%With the rapid development of e-commerce,online shopping becomes more and more appealing.Before shopping online, consumers usually tend to refer to the relevant comments to decide whether to buy the products or not.Therefore,to identify fake reviewers is of great significance.Based on the difference of emotional polarities between fake reviewers and real reviewers,we added the sentiment features of comment text to feature modelling process.Combined with the inter-relationship between users and specific commodities,we constructed a multi-edge graph model and came up with a method of spotting fake reviewers.Experimental results verified the effectiveness of the proposed algorithm.
展开▼
