【摘要】 随着Internet的飞速发展,Web应用系统在电子政务与电子商务中得到广泛应用,安全问题随之产生.入侵检测是保障Web应用系统安全的重要手段之一,利用可视化技术辅助安全专家创建轮廓有助于提高正常行为轮廓的准确程度,进而提高入侵检测性能.然而,传统基于散乱点的可视化模型对大样本数据的显示效果较差,在Internet环境中应用受限.本文针对传统模型的缺陷,提出了基于密度场的可视化模型及其相关算法,为安全专家提供更丰富的可视信息,以便安全专家能更准确地创建正常用户行为轮廓.本文还通过实验对两种可视化模型的显示效果进行了对比.更多还原

【Abstract】 Web applications, which are widely used to provide E-commerce and E-government services with the spread use of Internet, are faced with serious threats of security now. Intrusion detection was an effective way to secure web applications. Using visualization technique will facilitate security experts to create normal behavior profiles more exactly, and improve the detection performance of intrusion detection. However, traditional visualization model based on scatter points was not suitable to some web applications due to its poor display effect for volumes of samples. This paper proposes a new visualization model based on density field and corresponding algorithm in order to provide richer visual information for security experts and facilitate security experts to creating profiles more exactly. A comparison of experimental results between traditional model and new model was presented in this paper.更多还原