计算机工程 ›› 2019, Vol. 45 ›› Issue (8): 287-295.doi: 10.19678/j.issn.1000-3428.0051603

• 开发研究与工程应用 • 上一篇    下一篇

社交网络水军用户的动态行为分析及在线检测

李岩, 邓胜春, 林剑   

  1. 浙江财经大学 信息管理与工程学院, 杭州 310018
  • 收稿日期:2018-05-21 修回日期:2018-07-23 出版日期:2019-08-15 发布日期:2019-08-08
  • 作者简介:李岩(1992-),男,硕士研究生,主研方向为数据挖掘、社交网络;邓胜春,教授、博士;林剑,副教授、博士。
  • 基金项目:
    国家自然科学基金(61503331)。

Dynamic Behavior Analysis and Online Detection of Spammer User in Social Network

LI Yan, DENG Shengchun, LIN Jian   

  1. School of Information Management and Engineering, Zhejiang University of Finance and Economics, Hangzhou 310018, China
  • Received:2018-05-21 Revised:2018-07-23 Online:2019-08-15 Published:2019-08-08

摘要: 利用社交网络用户的静态行为特征识别水军用户,无法检测水军用户的动态行为且难以应用于在线检测的环境。为此,构造社交网络用户的动态行为特征,分析正常用户和水军用户间的差异,以半监督模型为基础,结合动静行为特征构建在线检测模型,通过静态行为特征聚类及动态行为特征过滤筛选,使半监督模型利用最有价值的未标记用户数据进行增量学习,从而检测水军用户。实验结果表明,该模型的F1值高达93.33%,平均训练时间约为2 min,能够有效检测社交网络上的水军用户。

关键词: 社交网络, 水军检测, 动态行为, 半监督模型, Tri-Training模型, 在线检测

Abstract: Using the static behavior characteristics of social network users to identify spammer users,it is impossible to detect the dynamic behavior of spammer users and it is difficult to apply to online detection.Therefore,by constructing the dynamic behavior characteristics of social network users,the differences between normal users and spammer users are analyzed.Based on the semi-supervised model,the online detection model is constructed by using dynamic and static behavior characteristics.The data is filtered by using static behavior characteristic clustering and dynamic behavior filtering.The semi-supervised model uses the most valuable unlabeled user data for incremental learning to detect spammer users.Experimental results show that the F1 value of the model is as high as 93.33% and the average training time is about 2 minutes,which can effectively detect the spammer users on social network.

Key words: social network, spammer detection, dynamic behavior, semi-supervised model, Tri-Training model, online detection

中图分类号: