Improving security using SVM-based anomaly detection: issues and challenges

摘要: Security is one of the main requirements current computer systems, and recently it gains much importance as number severity malicious attacks increase dramatically. Anomaly detection branches intrusion systems which enables to recognize newer variants security attacks. This paper focuses on anomaly schemes (ADS), have applied support vector machine (SVM) for detecting intrusions For this purpose, first presents required concepts about SVM classifier systems. It then classifies ADS approaches discusses various learning artificial intelligence techniques that been in combination with detect anomalies. Besides, specifies primary capabilities, possible limitations, or advantages approaches. Furthermore, a comparison studied provided illuminate their technical details.