Detection and defense of network virus using data mining technology

The spread of network viruses has posed a serious threat to the security network; therefore, it is necessary detect and defend them effectively. This paper used debugging application programming interface (API) technology obtain features API calls as viruses, filtered according information entropy, finally support vector machine (SVM) model for virus detection. experimental results showed that when number was 1200, algorithm had best detection performance, with an average true positive rate (TPR) 95.2%, false (FPR) 3.31%, overall accuracy 95.42%; compared C4.5 algorithm, K-means Naive Bayes SVM performance. show proposed method effective in defense can be further promoted applied practice.