N-gram based malicious code detection using Support Vector Machine learning approach

The development of Information and Communication gives a lot of convenience in our life, but on the other hand the new cyber threat such as viruses and computer intrusions also increases. This work discussed about the various classification algorithms such as K Nearest Neighbor (K-NN), Naive Bayes (NB) classifier and Support Vector Machine (SVM). Since it has been found that the SVM classification was efficient among the three classifiers as it can effectively perform linear classification and nonlinear classification of data, detects unknown and known malicious codes and has the capacity to handle large amount of features than the other two classifiers. This work focus on the detection of malicious code based on N-Gram and Support Vector Machine (SVM). It has been found that this approach can efficiently detect and classify the malicious data than other classifiers discussed here.