项目名称: 微博虚假信息及早检测与有效控制关键技术研究
项目编号: No.61202271
项目类型: 青年科学基金项目
立项/批准年度: 2013
项目学科: 计算机科学学科
项目作者: 谢柏林
作者单位: 广东外语外贸大学
项目金额: 23万元
中文摘要: 目前微博迅速发展,已成为人们获取信息和发布信息的一个重要平台。然而微博也正成为虚假信息滋生和泛滥的温床。为了减少微博虚假信息对公众的影响,使微博在人们的生产和生活中发挥更积极的作用,本项目拟对微博虚假信息及早检测与有效控制中的一些关键技术进行研究,主要研究内容如下:(1)研究一种基于隐半马尔可夫模型的微博信息流行度评估方法,用于识别微博上潜在的流行信息;(2)在微博信息流行度评估方法的基础上,研究一种基于隐半马尔可夫模型的微博虚假信息及早检测方法;(3)研究一种基于分类算法的微博信息发送者可信度评估方法,以便提高微博虚假信息及早检测方法的检测率;(4)为了净化微博信息的传播环境、抑制微博虚假信息的传播,研究一种基于社区检测算法的恶意转发者和恶意评论者检测方法,用于识别微博虚假信息传播过程中的恶意转发者和恶意评论者;(5)研究一种面向微博虚假信息的控制方法,用于有效抑制微博虚假信息的传播。
中文关键词: 微博;虚假信息;隐半马尔可夫模型;流行信息;可信度评估
英文摘要: In recent years, microblog has become a popular social medium for information dissemination. However, it could be seen that a lot of false information is sent and forwarded every day in microblog, due to the lack of an effective scheme for detection and countermeasures. In order to prevent the spread of false information and make microblog more user-friendly, this project proposes to research the critical issues related to the early detection of false information, and to develop the key techniques for effective controlling of the false information. The main research contents of this project are as follows:(1) This project intends to develop a popularity evaluation method for information based on hidden semi-Markov models, in order to identify the potentially popular information in microblog;(2) Based on the evaluation method and hidden semi-Markov models, this project intends to develop an early detection method for false information in microblog;(3) This project intends to develop a credibility evaluation method for information senders based on classification algorithms. The method could further improve the detection ratio of the early detection of false information;(4) In order to control the spread of false information in microblog, this project intends to develop a detection method for users who maliciously
英文关键词: microblog;misinformation;hidden semi-markov model;popular information;credibility evaluation