摘要
针对不良网络内容对数字教育的使用者尤其是青少年用户的威胁,提出了不良网络内容的识别与阻断系统模型,并研究了其中3项关键技术,包括高速网络流量数据的获取与还原技术、不良网络内容快速识别技术以及不良网络内容的自动阻断技术。在此基础上设计开发出一套不良网络内容识别与阻断系统,在Cernet西北网中心进行的验证与测试中,不良网络内容快速识别技术的准确率达到98.1%,不良网络内容自动阻断技术的成功率达到97.3%,为保障校园网的绿色安全提供了可靠的技术保证。
To solve the problem of harmful information spreading for e-Learning users, it is proposed a system frameworkfor harmful network content detection and blocking, and three key methods are studied, including capturing and recovering method for large scale network traffic data, harmful content detection method and automatic blocking method for harmful content visiting. Based on these methods, a system is developed and tested in the real environment of northwest center of Cemet. The results show that the accuracy rate of harmful content detection method is 98.1%, the success rate of blocking method for harmful content visiting is 97.3%, and the system could detect and block the harmful information spreading effectively.
基金
国家自然科学基金资助项目(60825202)
国家高技术研究发展计划(863计划)资助项目(2008AA01Z131)
国家科技支撑计划资助项目(2011BAK08B02)
关键词
不良内容识别
不良访问阻断
网络内容安全
harmful content detection
harmful visiting blocking
network content security