Visible to the public Biblio

Filters: Author is Ashiku, L.  [Clear All Filters]
A B C D E F G H I J K L M N O P Q R S T U V W X Y Z 
A
Ashiku, L., Dagli, C..  2020.  Agent Based Cybersecurity Model for Business Entity Risk Assessment. 2020 IEEE International Symposium on Systems Engineering (ISSE). :1—6.

Computer networks and surging advancements of innovative information technology construct a critical infrastructure for network transactions of business entities. Information exchange and data access though such infrastructure is scrutinized by adversaries for vulnerabilities that lead to cyber-attacks. This paper presents an agent-based system modelling to conceptualize and extract explicit and latent structure of the complex enterprise systems as well as human interactions within the system to determine common vulnerabilities of the entity. The model captures emergent behavior resulting from interactions of multiple network agents including the number of workstations, regular, administrator and third-party users, external and internal attacks, defense mechanisms for the network setting, and many other parameters. A risk-based approach to modelling cybersecurity of a business entity is utilized to derive the rate of attacks. A neural network model will generalize the type of attack based on network traffic features allowing dynamic state changes. Rules of engagement to generate self-organizing behavior will be leveraged to appoint a defense mechanism suitable for the attack-state of the model. The effectiveness of the model will be depicted by time-state chart that shows the number of affected assets for the different types of attacks triggered by the entity risk and the time it takes to revert into normal state. The model will also associate a relevant cost per incident occurrence that derives the need for enhancement of security solutions.