基于稀疏观测数据的网络恶意信息溯源与风险评估

Modern network is a double-edged sword, since its openness and sharing causes the massive and rapid spread of highly contagious malicious information. It is of great significance to identify the sources of malicious information and evaluate network users' security states, which can be used to combat the diffusers, repair network vulnerabilities, screen and protect users at high risk in a timely manner. However, current solutions face several challenges: 1) due to the cost and privacy protection reasons, the global observed information is extremely difficult to obtain; 2) they cannot be applied to dynamic topology of networks; 3) it is hard to deal with the cross-platform and multi-source environment; 4）the accuracy of risk evaluation for users is low. To end this, we first build a user-level dynamic diffusion model based on the sparse observed data for malicious information. On the basis of this diffusion model, we plan to realize the sources identification for cross-platform and dynamic networks with multiple sources. Finally, we will realize the real-time risk assessment for those unsupervised network users. In the context that network security has become more and more severe and a growing worldwide awareness of the need for strengthened privacy protection, this project is of great theoretical and practical value and is worthy of in-depth study.

现代网络是一把双刃剑，其开放性和共享性导致具有极强感染性的网络恶意信息迅速、大规模传播。对恶意信息进行溯源以及对用户的安全状态进行评估具有重大意义，可以用于打击恶意信息散布者、修复网络漏洞，并及时地甄别、保护处于高风险的用户节点。目前的解决方案存在如下挑战：1）由于成本和用户隐私保护因素，极难获取全网监测数据；2）不适合网络拓扑结构的动态变化；3）难以应对跨平台、多源头的传播环境；4）网络用户风险评估的准确率不高。本项目拟解决上述问题，首先基于稀疏观测的数据建立网络恶意信息的用户级的微观扩散模型；然后在此模型基础之上实现对跨多个网络平台的、动态的、多源头的网络恶意信息进行信息源头追溯；最后实现对网络中的未观测用户进行实时的风险评估。在网络信息安全形势越来越严峻和用户隐私保护意识逐渐加强的大背景下，本项目具有重大的理论与实践价值，值得深入开展研究。

现代网络是一把双刃剑，其开放性和共享性导致具有极强感染性的网络恶意信息迅速、大规模传播。对恶意信息进行溯源以及对用户的安全状态进行评估具有重大意义，可以用于打击恶意信息散布者、修复网络漏洞，并及时地甄别、保护处于高风险的用户节点。目前的解决方案存在如下挑战：1）由于成本和用户隐私保护因素，极难获取全网监测数据；2）不适合网络拓扑结构的动态变化；3）难以应对跨平台、多源头的传播环境；4）网络用户风险评估的准确率不高。本项目拟解决上述问题，首先基于稀疏观测的数据建立网络恶意信息的用户级的微观扩散模型；然后在此模型基础之上实现对跨多个网络平台的、动态的、多源头的网络恶意信息进行信息源头追溯；最后实现对网络中的未观测用户进行实时的风险评估。在网络信息安全形势越来越严峻和用户隐私保护意识逐渐加强的大背景下，本项目具有重大的理论与实践价值，值得深入开展研究。