期刊名称：International Journal of Security and Its Applications
印刷版ISSN：1738-9976
出版年度：2014
卷号：8
期号：5
页码：391-402
DOI：10.14257/ijsia.2014.8.5.34
出版社：SERSC
摘要：Internet worms are malware programs that imitate themselves and spread around the network. Internet worm, a wide spreading malcode exploits vulnerability in the operating system, hard disk, software and web browsers. This paper analyzes and classifies the Internet worm, depending on the training signatures. This work presents the Internet worm detection mechanism, using Principal Component Analysis (PCA) and Support Vector Machine (SVM). A Selective sampling technique is applied to maximize the performance of the classifier and to reduce misleading data instances. The results obtained show improved memory utilization, detection time and detection accuracy for Internet worms.
关键词：Malcode; Selective sampling; Multiclass SVM and PCA
