摘要
针对微博中谣言泛滥的问题,提出一种自动识别谣言的方法。该方法基于机器学习的原理,并在前人的基础上,结合赞的数目和置疑度两个新特征。实验结果显示结合新特征实现的系统在识别谣言上准确率达到82%,验证所提出的方法与特征的可行性和有效性。
Aiming at the spread of rumor in microblog system, proposes an automatic rumor detection method. It is based on the principle of machine learning and combined with the number of pros as well as the number of the doubt on the basis of previous studies. The experiment shows that system with new features reaches 82% accuracy rate. Thus, it proves that system that implemented is feasible and two new fea-tures are efficient.
出处
《现代计算机》
2016年第5期40-43,共4页
Modern Computer
基金
四川省科技厅项目(No.2014JY0036)
四川省教育厅创新团队基金(No.13TD0014)
关键词
谣言
社交网络
微博
机器学习
Rumor
Social Media
Microblog
Machine Learning
