File Relation Graph Based Malware Detection Using Label Propagation

摘要: The rapid development of malicious software programs has posed severe threats to Computer and Internet security. Therefore, it motivates anti-malware industry develop novel methods which are capable protecting users against new threats. Existing malware detectors mostly treat the file samples separately using supervised learning algorithms. However, ignoring relationship among limits capability detectors. In this paper, we present a detection method based on relation graph detect newly developed samples. When constructing graph, k-nearest neighbors chosen as adjacent nodes for each node. Files connected with edges represent similarity between corresponding nodes. Label propagation algorithm, propagates label information from labeled unlabeled files, is used learn probability that one unknown classified or benign. We evaluate effectiveness our proposed real large dataset. Experimental results demonstrate accuracy outperforms other existing approaches in classifying