Advancing Network Security: Enhancing Dynamic Vulnerability Detection in Secure and Insecure Programming through SDN-ML Hybrid Architecture
T. Sultan, Saloua Hendaoui · Research Square · 2023
Abstract Software-Defined Networking (SDN) is a new technique or method for managing computer networks. It contains several characteristics that allow the network to be set in a way that makes it easier to administer at a lower cost and improved efficiency to meet the demands of today's technology. However, three key factors must be assured while developing these networks: integrity, availability, and confidentiality. These networks are written in a variety of programming languages, including Java, and their frameworks may be checked and evaluated using software metrics to assure their integrity. The majority of software metrics tools primarily rely on static analysis and lack the capability to identify dynamic errors that arise from intricate abstract data structures like linked lists. Machine learning has the potential to enhance code analysis and identify vulnerabilities by addressing these dynamic challenges. This paper will suggest leveraging machine learning techniques to supplement software metrics tools, thereby addressing the limitation of static analysis in detecting dynamic errors originating from complex abstract data structures like linked lists. This novel approach aims to enhance code analysis and bolster the identification of vulnerabilities in software systems.