Virus Recognition Based on Combination of Hashing and Neural Networks

In this paper, we propose an intelligent first-warning system for virus code detection based on Artificial Neural Networks (ANNs). The proposed system operates in accordance with the basic principles of ANNs to conduct pattern matching of 32-bit hash signatures and detect virus signatures by means of the hashing applied to the byte content of executable code. The proposed system can accurately detect virus code in accordance with information it has learned, and gives false positive ratios within acceptable ranges. The results of experiments conducted show that the combination of 32-bit hashing and neural networks results in a low false positive rate. This paper also discusses the key ideas and approaches, along with the necessary adaptations and adjustments undertaken in the neural network model underlying the proposed early warning virus detection system..