Applications of spiking neural P systems in cybersecurity

Abstract

AbstractSpiking neural P systems are third-generation neural networks that are much more energy efficient than the current ones. In this paper, we investigate for the first time the possibility of using spiking neural P systems to solve cybersecurity-related problems. We proposed a new architecture called cyber spiking neural P systems (Cyber-SN P systems for short), which is designed especially for cybersecurity data and problems. We trained multiple Cyber-SN P systems to detect malware on the Android platform, phishing websites, and spam e-mails. We show through experiments that these networks can efficiently classify cybersecurity-related data with much fewer training epochs than perceptron-based artificial neural networks.