Trojan Detection Model of Nonlinear SVM Based on an Effective Feature Selection Optimization Algorithm

There are two major issues in the current Trojan detection system: some of them can not detect unknown Trojans and many of them have low detection rate. To solve these problems, a Trojan horse detection model of nonlinear SVM based on an effective feature selection optimization algorithm is presented in this paper. In this model, we extract the API (application program interface) calls sequence of an executable program as a feature vector and use the feature selection optimization algorithm to choose High-sensitive characteristics which are quantized into data recognized by SVM to build the SVM feature vector library. SVM classifier is trained with the training dataset to find the optimal separating hyper plane. Experiment results demonstrate that this model named PMI-SVM is more effective and steady.