Detection of advanced persistent threat using machine-learning correlation analysis

Abstract

Item Type Article Authors Ghafir, Ibrahim; Hammoudeh, M.; Prenosil, V.; Han, L.; Hegarty, R.; Rabie, K.; Aparicio-Navarro, F.J. Citation Ghafir I, Hammoudeh M, Prenosil V (et al) Detection of advanced persistent threat using machine-learning correlation analysis. Future Generation Computer Systems. 89: 349-359. Rights Citation: Ghafir I, Hammoudeh M, Prenosil V (et al) Detection of advanced persistent threat using machine-learning correlation analysis. Future Generation Computer Systems. 89: 349-359. AbstractAs one of the most serious types of cyber attack, Advanced Persistent Threats (APT) have caused major concerns on a global scale. APT refers to a persistent, multi-stage attack with the intention to compromise the system and gain information from the targeted system, which has the potential to cause significant damage and substantial financial loss. The accurate detection and prediction of APT is an ongoing challenge. This work proposes a novel machine learning-based system entitled MLAPT, which can accurately and rapidly detect and predict APT attacks in a systematic way. The MLAPT runs through three main phases: (1) Threat detection, in which eight methods have been developed to detect different techniques used during the various APT steps. The implementation and validation of these methods with real traffic is a significant contribution to the current body of research; (2) Alert correlation, in which a correlation framework is designed to link the outputs of the detection methods, aims to identify alerts that could be related and belong to a single APT scenario; and (3) Attack prediction, in which a machine learning-based prediction module is proposed based on the correlation framework output, to be used by the network security team to determine the probability of the early alerts to develop a complete APT attack. MLAPT is experimentally evaluated and the presented system is able to predict APT in its early steps with a prediction accuracy of 84.8%.to a technical report by Trend Micro [11], this assumption is no longer valid with the rise of targeted attacks, Advanced Persistent Threats (APTs), in which both cyber-criminals and hackers are targeting selected organizations and persisting until they achieve their goals.The APT attack is a persistent, targeted attack on a specific organisation 20 and is performed through several steps [12]. The main aim of APT is espionage and then data exfiltration. Therefore, APT is considered as a new and more complex version of multi-step attack. These APTs present a challenge for current detection methods as they use advanced techniques and make use of unknown vulnerabilities. Moreover, the economic damage due to a successful 25 APT attack significant. The potential cost of attacks is the major motivation for the investments in intrusion detection and prevention systems [13]. APTs are currently one of the most serious threats to companies and governments [14].Most of the research in the area of APT detection, has focused on analysing already identified APTs [15][16][17][...

Artificial Intelligence in the Cyber Domain: Offense and Defense

“…Another approach [53] introduced a ML-based approach named MLAPT to identify and predict APTs. According to the authors, their system had the ability of early prediction of APT attacks.…”

A Fuzzy WASPAS-Based Approach to Determine Critical Information Infrastructures of EU Sustainable Development

“…They often lead to numerous and significant losses, the essential disruption of production, and the destruction of the environment, etc. [11][12][13][14].…”

Addressing Multi-Stage Attacks Using Expert Knowledge and Contextual Information

“…In particular, this IDS focuses on detecting a 5-steps MSA, in real-time, without previous training process. The main goal of this MSA is to create a Point of Entry (PoE) to a targeted machine, which could be used for the completion of an Advanced Persistent Threat (APT) like attack [6]. As we demonstrated in [7], a Fuzzy Cognitive Map (FCM) [8] can be used to incorporate the PoL into the detection process.…”