基于DCA算法的微博虚假信息检测  

Micro-blog Misinformation Detection Based on Dendritic Cells Algorithm

作　　者：杨超 项振辉 李涛[3] Yang Chao;Xiang Zhenhui;Li Tao(School of Computer Science and Information Engineering,Hubei University,Wuhan430062,China;Hubei Education Information Engineering Technology Research Center,Wuhan430062,China;Hubei Provincial Key Lab of Intelligent Information Processing and Real-Time Industrial Systems,Wuhan University of Science and Technology,Wuhan430065,China)

机构地区：[1]湖北大学计算机与信息工程学院,武汉430062 [2]湖北省教育信息化工程技术研究中心,武汉430062 [3]武汉科技大学智能信息处理与实时工业系统湖北省重点实验室,武汉430065

出　　处：《计算机测量与控制》2019年第3期183-187,191,共6页Computer Measurement &Control

基　　金：国家自然科学基金(61170306);智能信息处理与实时工业系统湖北省重点实验室开放基金项目(znxx2018MS05)

摘　　要：通过对微博虚假信息的分析,基于DCA算法的思想,提出一种检测微博虚假信息的方法;以新浪微博为例,从虚假信息发布者的用户属性和虚假信息评论的文本内容两个方面进行分析;从用户方面选取用户的特征属性,如是否认证、有无简介、地址信息是否详细、关注数、粉丝数等,从评论内容方面选取评论与微博内容的相关性、评论的支持性及其置信度等特征属性,将以上属性的分析结果作为区别虚假信息与真实信息的特征信号,并基于树突状细胞算法(Dendritic Cells Algorithm,DCA)实现新浪微博虚假信息的识别;使用新浪微博真实数据对算法有效性进行了验证和对比实验,结果表明该方法能够有效检测出新浪微博中的虚假信息,具有较高的检测准确率。By the comprehensive analyzing of misinformation on micro-blog, applied the idea of Dendritic Cells Algorithm provided a method to test the misinformation on micro-blog, took the Sina micro-blog as an example, analyzed the information from two sides, one side was the property of misinformation publisher, the other side was the comment content of the misinformation, selected the user’s properties, such as, identified whether the publisher was certified, had brief introduction, had detailed address, number of attention, number of fans etc. selected the comments’ properties,such as the relativity between the content of micro-blog and comment, the support and confidence of comment etc. took the above property analyzing results as property signals to identify the misinformation and real information, and realized the identification of false information based on the Dendritic Cells Algorithm( DCA) on Sina micro-blog. We did the comparison experiment and test the effectiveness of the algorithm by using the real data on Sina micro-blog, the result shows that this method can test the false information on Sina micro-blog effectively, has high test accuracy rate.

关 键 词：微博 虚假信息 人工免疫 危险理论 树突状细胞算法 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]