计算机应用 ›› 2017, Vol. 37 ›› Issue (10): 2799-2805.DOI: 10.11772/j.issn.1001-9081.2017.10.2799

• 网络空间安全 • 上一篇    下一篇

基于突发话题和领域专家的微博谣言检测方法

杨文太1, 梁刚2, 谢凯1, 杨进2, 许春2   

  1. 1. 四川大学 计算机学院, 成都 610065;
    2. 四川大学 网络空间安全学院, 成都 610065
  • 收稿日期:2017-04-28 修回日期:2017-07-24 出版日期:2017-10-10 发布日期:2017-10-16
  • 通讯作者: 梁刚(1976-),男,四川成都人,副教授,博士,主要研究方向:网络安全、智能计算,E-mail:lianggang@scu.edu.cn
  • 作者简介:杨文太(1993-),男,甘肃庆阳人,硕士研究生,主要研究方向:网络安全、谣言检测;梁刚(1976-),男,四川成都人,副教授,博士,主要研究方向:网络安全、智能计算;谢凯(1992-),男,四川成都人,硕士研究生,主要研究方向:网络安全、舆情监测;杨进(1980-),男,四川乐山人,副研究员,博士,主要研究方向:网络安全、智能计算;许春(1972-),男,河北石家庄人,副教授,博士,主要研究方向:网络安全、智能计算.
  • 基金资助:
    四川省教育厅重点资助项目(17ZA0238,17ZA0200);四川省学术和技术带头人培养支持经费资助项目(2016)。

Rumor detection method based on burst topic detection and domain expert discovery

YANG Wentai1, LIANG Gang2, XIE Kai1, YANG Jin2, XU Chun2   

  1. 1. College of Computer Science, Sichuan University, Sichuan Chengdu 610065, China;
    2. College of Cyber Space Security, Sichuan University, Sichuan Chengdu 610065, China
  • Received:2017-04-28 Revised:2017-07-24 Online:2017-10-10 Published:2017-10-16
  • Supported by:
    This work is partially supported by the Research Foundation of Education Bureau of Sichuan Province (17ZA0238, 17ZA0200), the Sichuan Training Support Fund for Academic and Technical Leaders (2016).

摘要: 针对现有谣言检测方法中存在的数据采集困难和谣言检测滞后的问题,提出一种基于动量模型的突发话题检测和领域专家发现的谣言检测方法。该方法借鉴物理学中的动力学理论对话题特征进行建模,使用特征的动力学物理量描述特征的突发特性和发展趋势,并在对突发特征进行特征聚合之后提取得到突发话题;然后,依据话题与用户个人信息的领域相关性在候选专家池中发现领域相关的微博用户来甄别话题信息的真实性。基于新浪微博数据的实验结果表明,相对于仅基于有监督机器学习的微博谣言识别方法,该方法谣言识别准确率提高了13个百分点;相对于主流人工识别方法,将最长谣言检测用时缩短至20h,能够较好地应用于实际的微博谣言检测环境。

关键词: 动量模型, 话题, 突发, 领域专家, 谣言检测

Abstract: It is difficult for existing rumor detection methods to overcome the disadvantage of data collection and detection delay. To resolve this problem, a rumor detection method based on burst topic detection inspired by the momentum model and domain expert discovery was proposed. The dynamics theory in physics was introduced to model the topic features spreading among the Weibo platform, and dynamic physical quantities of the topic features were used to describe the burst characteristics and tendency of topic development. Then, emergent topics were extracted after feature clustering. Next, according to the domain relativity between the topic and the expert, domain experts for each emergent topic were selected within experts pool, which is responsible for identifying the credibility of the emergent topic. The experimental results show that the proposed method gets 13 percentage points improvement on accuracy comparing with the Weibo rumor identification method based merely on supervised machine learning, and the detection time is reduced to 20 hours compared with dominating manual methods, which means that the proposed method is applicable for real rumor detection situation.

Key words: momentum model, topic, burst, domain expert, rumor detection

中图分类号: