Script Virus Detection Algorithm Based on Fusion of Fuzzy Pattern and Decision Tree

The method using the decision tree for script virus detection can make full use of the information of training samples. But complex sample features and large number of samples will produce large number of nodes which result in the high algorithm time complexity and affect the classification accuracy due to the pruning process. In order to improve classification performance, a fusion algorithm using the information of fuzzy pattern is designed based on the decision tree classification algorithm. Three important characteristics of fuzzy pattern about close degree are regarded as the three attributes of sample information vector in the decision tree to build decision tree through training get. The stability and accuracy of the algorithm is verified by experiment. The experiment results show that the proposed algorithm increases discrimination of attributes and reduces the decision tree branch.