校园网智能恶意软件数据检测研究

Research on Malicious Software Data Detection of Campus Network

随着互联网技术的不断发展,恶意软件的几何式增长和传播,传统安全软件鉴别模式已经很难适应新形势,大量的未知文件经过杀毒软件比对后,仍有很高比例无法识别。通过数据挖掘技术实现对未知文件高效、准确识别、分析和处理就有较大的实用意义。采用分类关联规则挖掘算法建立模型,并使用集成学习方法共同进行未知文件预测,对预测的恶意软件使用聚类进行归类,并提取特征代码用于鉴别。

With the continuous development of the Intemet technology, the size of malicious software is growing and spreading at an incredible rate, so the traditional security software authentication mode can not adapt to the new situation. After using anti-virus software comparison, there is still a high proportion of unrecognized. Data mining technology to implement unknown files is efficient and accurate to identify and analyze for these files, so it has great practical significance. In this paper, the model is to built by classified association rule mining algorithm and integrated learning methods to carry out unknown files prediction, the prediction of malware uses cluster analysis to classify and extract the feature code for its identification.