Finding the Hidden Hands:A Case Study of Detecting Organized Posters and Promoters in SINA Weibo  

作　　者：WANG Xiang ZHANG Zhilin YU Xiang JIA Yan ZHOU Bin LI Shasha 

机构地区：[1]School of Computer,National University of Defense Technology [2]School of Computing Science,Simon Fraser University [3]State Key Laboratory of High Performance Computing,National University of Defense Technology

出　　处：《China Communications》2015年第11期143-155,共13页中国通信（英文版）

基　　金：supported by 973 Program of China(Grant No.2013CB329601, 2013CB329602,2013CB329604);NSFC of China(Grant No.60933005,91124002);863 Program of China(Grant No.2012AA01A401, 2012AA01A402);National Key Technology RD Program of China(Grant No.2012BAH38B04, 2012BAH38B06)

摘　　要：With the development of online social networks,a special group of online users named organized posters(or Internet water army,Internet paid posters in some literatures) have fl ooded the social network communities. They are organized in groups to post with specific purposes and sometimes even confuse or mislead normal users.In this paper,we study the individual and group characteristics of organized posters. A classifier is constructed based on the individual and group characteristics to detect them. Extensive experimental results on three real datasets demonstrate that our method based on individual and group characteristics using SVM model(IGCSVM) is effective in detecting organized posters and better than existing methods. We take a first look at finding the promoters based on the detected organized posters of our IGCSVM method. Our experiments show that it is effective in detecting promoters.With the development of online social networks,a special group of online users named organized posters(or Internet water army,Internet paid posters in some literatures) have fl ooded the social network communities. They are organized in groups to post with specific purposes and sometimes even confuse or mislead normal users.In this paper,we study the individual and group characteristics of organized posters. A classifier is constructed based on the individual and group characteristics to detect them. Extensive experimental results on three real datasets demonstrate that our method based on individual and group characteristics using SVM model(IGCSVM) is effective in detecting organized posters and better than existing methods. We take a first look at finding the promoters based on the detected organized posters of our IGCSVM method. Our experiments show that it is effective in detecting promoters.

关 键 词：organized posters internet water army online paid posters promoter MICROBLOGGING 

分 类 号：TP393.092[自动化与计算机技术—计算机应用技术]