作者: Yihua Liao , V. Rao Vemuri , Wenjie Hu
DOI:
关键词:
摘要: Using the 1998 DARPA BSM data set collected at MIT’s Lincoln Labs to study intrusion detection systems, performance of robust support vector machines (RVSMs) was compared with that conventional and nearest neighbor classifiers in separating normal usage profiles from intrusive computer programs. The results indicate superiority RSVMs not only terms high accuracy low false positives but also their generalization ability presence noise running time. Keywords—Intrusion detection, security, machines, noisy data.