Hybrid Swarm Intelligence and Artificial Neural Network for Mitigating Malware Effects
Tarek S. Sobh
Affiliation: Information Systems Department, Egyptian Armed Forces, Egypt, Address: 110 Zhraa Nasr City, Stage 1, Cairo, Egypt.
Keywords: Artificial neural network, local victim information, particle swarm optimization, swarm intelligence, worm detection,
Today networks are interconnected wired and wireless network. With the explosive growth and increasing
complexity of network applications, malware attacks such as worm attack against network are critical. Although of the
evolution of worm detection techniques, worms are still the most malware threats attacking computer systems. Early detection
of unknown worms is still a problem. Swarm Intelligence (SI) in recent patents seeks inspiration in the behavior of
swarms of insects or other animals such as ants. SI is applied in other fields with success. We used it in the field of worm
detection. Artificial neural networks may either be used to gain an understanding of biological neural networks, or for
solving artificial intelligence problems without necessarily creating a model of a real biological system.
This paper introduces a system for detecting unknown worms based on the collected information from local victim using
Particle Swarm Optimization (PSO) and Artificial Neural Network (ANN). This system can detect unknown worms effectively
in both small and large size networks. In addition, this system produces prediction to the infection percentage in the
network. This prediction mechanism supports the network administrator in decision-making process to respond quickly to
worm propagation accurately.
Rights & PermissionsPrintExport