面向社会安全事件的网络信息分析及态势预测研究

项目名称： 面向社会安全事件的网络信息分析及态势预测研究

项目编号： No.61271316

项目类型： 面上项目

立项/批准年度： 2013

项目学科： 无线电电子学、电信技术

项目作者： 李生红

作者单位： 上海交通大学

项目金额： 70万元

中文摘要： 当今高风险社会环境下，各种社会安全事件频发。随着互联网普及使用，社会安全事件常在其上有所反映，并且网络也对其起到了推波助澜的作用。通过分析网络信息以便为可能发生的社会安全事件提供预警，已成为迫切有待解决的问题。然而，由于该方面起步较晚，且其涉及的一些核心理论/技术尚未很好解决，目前这方面成果相对较少且性能存在明显不足。鉴于此，并考虑到社会安全事件在网上的主要反映常体现在相关媒体热/焦点和相应网络群体形成聚集，项目以面向社会安全事件的网络信息分析及态势预测新方案/方法为总体研究目标，具体研究网络文本特征提取表达、网络文本分/聚类、网络群体挖掘、媒体聚集及群体状态评估、网络安全事件态势预测等核心问题的有效解决方法，并进而研发具有网络安全事件预警等功能的相应网络信息分析及态势预测平台。预期成果可望对内容安全等相关领域的研究在理论上起到促进作用，也可望在网络信息管控方面具有良好应用前景。

中文关键词： 网络媒体信息分析；分类/聚类；学习自动机；群体结构挖掘；安全态势预测

英文摘要： Nowadays, under the high-risk social circumstance, various kinds of social security events occur frequently. With the popularity of Internet, more and more social security events can be traced on the Internet. Furthermore, the Internet can lead to the spread of the social security events. Analyzing the information from Internet in order to provide early warning of the possibly coming social security events has become an urgent issue to be solved. However, because the above issue has just been studied recently and some related critical theories/techniques have not been well addressed, there are only a little research outcome with limited performance by now. In view of the above, and considering that the social security events are usually reflected by the focus of related media information and the aggregation of the related network colonies on the Internet, the aim of the project is to study new Internet information analysis and situation forecasting schemes/methods related to the social security events. The major research include: network text feature extraction and representation, network text classification/clustering, network colony structure mining, media information aggregation and colony status evaluation, and situation forecasting of network security events, etc., furthermore, based on the above research o

英文关键词： Network media information analysis；Classification and clustering；Learning automata；Colony/ community structure mining；Security situation forecasting