Malware propagation modeling by the means of genetic algorithms

Existing malware propagation models mainly concentrate to forecasting the number of infected computers in the initial propagation phase. In this article we propose a genetic algorithm based model for estimating the propagation rates of known and perspective Internet worms after their propagation reaches the satiation phase. Estimation algorithm is based on the known worms’ propagation strategies with correlated propagation rates analysis and is presented as a decision tree, generated by GAtree v.2 application. Genetic algorithm approach for decision tree generation is selected taking into consideration the efficiency of this method while solving optimization and modeling tasks with large solution space. The performed tests have shown that the proposed model is efficient and can be used as a framework for modeling propagation rates after the satiation phase of different malware types.