🎓 Pendidikan
💼 Pengalaman Profesional
📚 Publikasi & Research
Machine Learning-Based Android Malware Detection: A 5-Year Retrospective
Comprehensive analysis of ML-based Android malware detection approaches from 2018-2023. We evaluated 50+ published methods across 10,000+ malware samples and identified key limitations and future research directions. Our findings show that ensemble methods combining static+dynamic analysis achieve 99.2% detection rate with 0.3% false positive rate.
Read Paper →PermGuard: Permission-based Android Security Framework for Zero-day Threat Detection
We propose PermGuard, a novel permission-based security framework that analyzes Android app behavior patterns to detect zero-day threats. Evaluated on 50,000+ apps, PermGuard achieved 97.8% detection rate for previously unseen malware with minimal performance overhead (< 2% CPU). The framework has been deployed in production by 3 major Android OEMs.
Read Article →Breaking Android: Advanced Exploitation Techniques for Security Researchers
Live demonstration of advanced Android exploitation techniques including kernel exploitation, SELinux bypass, and persistent backdoor installation. Showcased 5 zero-day vulnerabilities discovered during security audit of popular Android apps. All vulnerabilities were responsibly disclosed and patched before presentation.
Watch Talk →Android Malware Evolution: A Comprehensive Study of 100,000 Samples (2015-2020)
Longitudinal study analyzing evolution of Android malware over 5 years. We identified 7 major malware families, tracked their development patterns, and predicted future trends. Dataset of 100,000+ malware samples made public for research community. Study cited 150+ times in security research papers.
Read Paper →
🏆 Sertifikasi Profesional
