This repository contains the multi-view feature data and labels for Android Malware Capability Identification.
In the 'Botnet Attack Capability' folder, there are training and testing feature matrices and labels for the Botnet Attack Capability. These features are extracted from three views, including source-code view, smali-opcode view and API call graph view. Label '1' denotes malware sample with the Botnet Attack Capability while '-1' means the sample without the Botnet Attack Capability.
In the 'Information Stealing Capability' folder, there are training and testing feature matrices and labels for the Information Stealing Capability. These features are extracted from three views, including source-code view, smali-opcode view and API call graph view. Label '1' denotes malware sample with the Information Stealing Capability while '-1' means the sample without the Information Stealing Capability.
In the 'Unauthorized Root Access Capability' folder, there are training and testing feature matrices and labels for the Unauthorized Root Access Capability. These features are extracted from three views, including source-code view, smali-opcode view and API call graph view. Label '1' denotes malware sample with the Unauthorized Root Access Capability while '-1' means the sample without the Unauthorized Root Access Capability.
In the 'Malicious SMS Charge Capability' folder, there are training and testing feature matrices and labels for the Malicious SMS Charge Capability. These features are extracted from three views, including source-code view, smali-opcode view and API call graph view. Label '1' denotes malware sample with the Malicious SMS Charge Capability while '-1' means the sample without the Malicious SMS Charge Capability.
In case of any queries, please feel free to email: junzhang@swin.edu.au