Proposal / Submission Type
Peer Reviewed Paper
Start Date
17-5-2018 3:20 PM
End Date
17-5-2018 3:55 PM
Abstract
Enterprises are facing an unprecedented risk of security incidents due to the influx of emerging technologies, like smartphones and wearables. Most of the current Mobile security systems are not maturing in pace with technological advances. They lack the ability to learn and adapt from the past knowledge base. In the case of a security incident, enterprises find themselves underprepared for the lack of evidence and data. The systems are not designed to be forensic ready. There is a need for automated security analysis and forensically ready solution, which can learn and continuously adapt to new challenges, improve efficiency and productivity of the system. In this research, the authors have designed a security analysis and digital forensic readiness system targeted at smartphones and wearables in an enterprise environment. The proposed system detects applications violating security policies, analyzes Android and iOS applications to identify possible vulnerabilities on the server, apply machine learning algorithms to improve the efficiency and accuracy of vulnerability prediction. The System continuously learns from past incidents, proactively collect required information from the devices which can help in digital forensics. Machine learning techniques are applied to the set of features extracted from the decompiled Mobile applications and applications classified based on consisting of one or more vulnerabilities. The system was evaluated in a real-world enterprise environment with 14151 mobile applications and vulnerabilities was predicted with an accuracy of 94.2%. The system can also work on virtual instances of the mobile devices.
Scholarly Commons Citation
Govindaraj, Jayaprakash; Verma, Robin; and Gupta, Gaurav, "Precognition: Automated Digital Forensic Readiness System for Mobile Computing Devices in Enterprises" (2018). Annual ADFSL Conference on Digital Forensics, Security and Law. 11.
Digital Forensic Readiness, Machine Learning, APK analysis, IPA analysis, Mobile Application security analysis
Final edit - full res
Included in
Computer and Systems Architecture Commons, Forensic Science and Technology Commons, Information Security Commons
Precognition: Automated Digital Forensic Readiness System for Mobile Computing Devices in Enterprises
Enterprises are facing an unprecedented risk of security incidents due to the influx of emerging technologies, like smartphones and wearables. Most of the current Mobile security systems are not maturing in pace with technological advances. They lack the ability to learn and adapt from the past knowledge base. In the case of a security incident, enterprises find themselves underprepared for the lack of evidence and data. The systems are not designed to be forensic ready. There is a need for automated security analysis and forensically ready solution, which can learn and continuously adapt to new challenges, improve efficiency and productivity of the system. In this research, the authors have designed a security analysis and digital forensic readiness system targeted at smartphones and wearables in an enterprise environment. The proposed system detects applications violating security policies, analyzes Android and iOS applications to identify possible vulnerabilities on the server, apply machine learning algorithms to improve the efficiency and accuracy of vulnerability prediction. The System continuously learns from past incidents, proactively collect required information from the devices which can help in digital forensics. Machine learning techniques are applied to the set of features extracted from the decompiled Mobile applications and applications classified based on consisting of one or more vulnerabilities. The system was evaluated in a real-world enterprise environment with 14151 mobile applications and vulnerabilities was predicted with an accuracy of 94.2%. The system can also work on virtual instances of the mobile devices.
Comments
Visit the Panel Session page