A hybrid automatic intrusion detection system using machine learning technique to detect anomalous traffic for network security

Abstract

Secured provision of computer network in the fields of electronic newlinecommerce, government other critical service organizations are becoming newlinechallenging with every passing day. Security threats in the Internet are posing newlinea huge challenge in the recent day, world-wide. Hence, in the present era of newlineinformation technology computer / cyber security has become a high-priority newlineglobal issue that needs to be addressed. Networked computing which is an newlineinevitable part information system has made it vulnerable to security threats. newlineIntrusions compromise Confidentiality, Integrity and Availability (CIA) of newlinecomputing resources and data available in a networked environment, resulting newlinein heavy loss both in terms of money and trust to commercial or government newlineorganizations Thus it has become both mandatory and urgent for all the computer newlinenetworks to be guarded with multilevel security systems. Multilevel security newlinecan be provided with sophisticated software and equipments such as firewall, newlineVirtual Private Network (VPN), web and email filtering, antivirus protection, newlineevent management and vulnerability scanning tools. Most of the prevention newlinemethods just discussed is inadequate; there is a demanding need for a security newlinecompromise monitoring system. One such security breach monitoring system newlineis the Intrusion Detection System (IDS). Early and effective detection of newlineintrusions continues to be a big challenge to the automated IDS. Accurate newlinedetection of intrusion with less false positives (alarm without a security newlineincident) has been elusive as always.. newline newline