5Detection Using Graph Centrality Measures
In the Chapter 4, we showed that TAN model can be used to combine the static and dynamic features for accurate malware detection. In TAN-based detection model, it is possible for a malware developer to evade API call and permission-based classifiers by employing adversarial attacks[126].
Adversarial attacks on Android malware detection mechanisms are really threatening. They can make an attacker to gain unauthorized access to a device. In ML-based malware detection, feature values are extremely important since a slight change in the feature values can affect the output of the classifier. In addition to that, many adversarial attacks are transferable. That is the attacks ...
Get Intelligent Mobile Malware Detection now with the O’Reilly learning platform.
O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.
