A Detection Method Based on Clustering for Review Spam
There are more and more valuable reviews on products on the Internet Inevitably, there are some review spam such as advertisements and unrelated items in normal reviews. It is very necessary to propose one method to detect review spam so as to avoid interference when we get useful information from reviews. In this paper, we proposed one unsupervised detection method based on clustering for review spam. This experiment results show that this method has high precision and recall rate, and achieves preferable effects.
cluster review spam review analysis resistance distance
Hong Zhu Wentang Tan Mingke He Bin Ge Fangfang Li
Science and Technology on Information Systems Engineering Laboratory National University of Defense Technology Changsha, China
国际会议
海口
英文
194-197
2011-07-15（万方平台首次上网日期，不代表论文的发表时间）