Spying identification is an emerging field of software development and research and networking with increasing Internet use in daily life. There have been many issues with classical IDS systems, along with other low network attack identification, high false alarm rate, and inadequate analytical capacity. The primary potential of research on the subject then is to establish a model for intrusion detection with increased performance and decreased preparation time. Machine learning is an efficient tool for analyzing any abnormal events taking place in the network activity flow. This paper proposes a mixture of two methodologies to determine any abnormal conduct in internet traffic. This paper recommends to use the principal component analysis (PCA) and Rough-set Support Vector Machine (SVM) as the hybrid intrusion detection models.
Amareshwari, N. R., Ramanjaneyulu, S., & Swapna, G. (2019). Strategies for network intrusion detection using machine learning algorithms. International Journal of Recent Technology and Engineering, 8(1), 2437–2440.