AndroParse - An Android Feature Extraction Framework and Dataset

摘要: Android malware has become a major challenge. As consequence, practitioners and researchers spend significant time analyzing applications (APK). A common procedure (especially for data scientists) is to extract features such as permissions, APIs or strings which can then be analyzed. Current state of the art tools have three issues: (1) single tool cannot all used by scientists (2) are not designed extensible (3) Existing parsers timely they runtime efficient scalable. Therefore, this work presents AndroParse an open-source parser written in Golang that currently extracts four most features: Permissions, APIs, Strings Intents. outputs JSON files easily programming languages. Constructing allowed us create extensive feature dataset accessed our independent REST API. Our 67,703 benign 46,683 malicious APK samples.