A malware detection model based on improved Harris hawks algorithm to optimize support vector machine SVM and feature selection was proposed.To improve the ability of feature subset selection and the classification accuracy of support vector machine,the chaotic mapping,nonlinear periodic adjustment of energy factor,optimal solution variation disturbance and mutually beneficial symbiosis strategy were used to improve the initial population structure,global search and local mining switching per-formance and jumping off local optimization ability of the HHO algorithm.The improved HHO was used to synchronously opti-mize the SVM parameter optimization and feature subset selection.A malware detection model was constructed.The results show that the improved algorithm can achieve higher classification accuracy while reducing the feature dimension,high-quality feature subsets are used to improve the classification ability of malware detection model.
维普
万方数据