Cyber Supply Chain Threat Analysis and Prediction Using Machine Learning and Ontology

Abstract

Cyber Supply Chain (CSC) security requires a secure integrated network among the sub-systems of the inbound and outbound chains. Adversaries are deploying various penetration and manipulation attacks on an CSC integrated network's node. The different levels of integrations and inherent system complexities pose potential vulnerabilities and attacks that may cascade to other parts of the supply chain system. Thus, it has become imperative to implement systematic threats analyses and predication within the CSC domain to improve the overall security posture. This paper presents a unique approach that advances the current state of the art on CSC threat analysis and prediction by combining work from three areas: Cyber Threat Intelligence (CTI ), Ontologies, and Machine Learning (ML). The outcome of our work shows that the conceptualization of cybersecurity using ontological theory provides clear mechanisms for understanding the correlation between the CSC security domain and enables the mapping of the ML prediction with 80% accuracy of potential cyberattacks and possible countermeasures.

The Diversification and Enhancement of an IDS Scheme for the Cybersecurity Needs of Modern Supply Chains

“…Common risks are data breaches and corporate fraud originating from ransomware attacks and malware infections [8]. A solution to these issues is the use of tools that promptly identify and provide alerts about malicious activity in a supply chain infrastructure and, subsequently, limit the impact of the attack [9]. Such type of tools are the intrusion detection systems (IDSs) [10][11][12] and the intrusion prevention systems (IPSs) [13,14].…”

Cyber Threat Ontology and Adversarial Machine Learning Attacks: Analysis and Prediction Perturbance

“…Biggio et al 2011 explore adversarial data manipulations using SVM classification algorithm under adversarial label noise by subverting the SVM learning process [20]. Chen et al 2017 designed a randomizing SVM model by using robust SVMs Gaussian distribution against method for adversarial attacks [26]. Munoz-Gonzalez et al 2017 proposed a novel poisoning algorithm based on a back-gradient optimization and used neural networks algorithms and deep learning architectures techniques to learn the dataset.…”

“…The overall features were 64. We extracted 38 features in the primary data relevant to the attack profile [26] [27].…”

Cyberattack Ontology: A Knowledge Representation for Cyber Supply Chain Security

“…Cyberattack ontology from the CSC perspective describes organizational security concepts, properties relationships and their interdependencies in a formal and structured approach for analysis and intelligence gatherings. [7] [15]. The goal of the cyberattack ontology is to extract relevant attack instances and information from data to ensure consistency and accuracy in the CSC system security concepts and for knowledge reuse in the threat intelligence domain.…”

“…Additionally, due to the varying organizational goals and dynamic requirements, various integrations, varying business processes, and delivery mechanisms, predicting cyberattacks in CSC from an organizational perspective has been challenging. To address these challenges, we consider the CSC cyberattack modelling approach using ontology concepts for knowledge representation and reuse within the CSC domain [15] [16] as shown in Figure 1.…”