AndroMetric: Bridging Multi-Dimensional Software Metrics and Mobile Application Security
Year: 2026
Type: conference-paper
Abstract: Android applications are widely used and increasingly targeted by malware authors, posing ongoing challenges for security research and app marketplace vetting. Understanding how internal software characteristics relate to app security is a growing research need, yet large-scale datasets linking code metrics with malicious behavior remain scarce. This paper introduces AndroMetric, a dataset designed to bridge software metrics and mobile application security by capturing multi-dimensional structural characteristics and statically inferred behavioral indicators across benign and malicious Android apps. The dataset includes 31 static code metrics extracted from 39,243 APKs (29,208 malware samples and 10,035 benign apps), along with other metadata. Metrics span six categories: size and dimensional structure, object-oriented design, control-flow complexity, sensitive operation usage, Android Intent patterns, and exception handling. AndroMetric supports diverse research applications including malware detection, code quality comparison, and mining trends in Android malware development.
Authors: Sebastian Siedler, Karim Elish
Institution: Florida Polytechnic University
Language: English
Subfield: Software
Field: Computer Science
Domain: Physical Sciences
Open Access status: hybrid
OpenAlex ID: W7172034485