Malware Detection with Deep Neural Network Using Process Behavior 论文

摘要

Increase of malware and advanced cyber-attacks are now becoming a serious problem. Unknown malware which has not determined by security vendors is often used in these attacks, and it is becoming difficult to protect terminals from their infection. Therefore, a countermeasure for after infection is required. There are some malware infection detection methods which focus on the traffic data comes from malware. However, it is difficult to perfectly detect infection only using traffic data because it imitates benign traffic. In this paper, we propose malware process detection method based on process behavior in possible infected terminals. In proposal, we investigated stepwise application of Deep Neural Networks to classify malware process. First, we train the Recurrent Neural Network (RNN) to extract features of process behavior. Second, we train the Convolutional Neural Network (CNN) to classify feature images which are generated by the extracted features from the trained RNN. The evaluation result in several image size by comparing the AUC of obtained ROC curves and we obtained AUC= 0:96 in best case.