Title: Innovative Architectural Framework Design for an Effective Machine Learning Based APT Detection

Issue Number: Vol. 11, No. 1
Year of Publication: 2021
Page Numbers: 12-22
Authors: Mourad M.H Henchiri, Sharyar Wani
Journal Name: International Journal of Digital Information and Wireless Communications (IJDIWC)
- Hong Kong

Abstract:


Generating regular rules to be passed to different security appliances set within the work environment, is a stressing job to be carefully set; since even the data bouncing and the data collision impact over a network trunk might be considered as an anomaly by a variety of filters [45, 47, 48, 49, 52]. Network data flows load balancing, also, leads to a pattern based anomaly when not configured as per the network potential [46, 48, 49]. Thus, securing a platform against the APT attacks, whether a prevention scenario or a detection process, research demonstrates that security intelligence and big data analytics would enormously prevent and detect abnormalities, this is all by keeping an eye on the difficulty of data classification [44, 50, 51, 52, 53, 56]. In this research we would be generating an APT detection framework diagram via which we enhance the weaknesses seen in regular and commercialized filter based IDS, IPS and Firewalls. Which would give a remarkably enhanced live data flow clustering and classification algorithm.