악성 안드로이드 앱 탐지를 위한 개선된 특성 선택 모델 = Advanced Feature Selection Method on Android Malware Detection by Machine Learning

다국어 초록 (Multilingual Abstract)

According to Symantec's 2018 internet security threat report, The number of new mobile malware variants increased by54 percent in 2017, as compared to 2016. And last year, there were an average of 24,000 malicious mobile applicationsblocked each day. Existing signature-based technologies of malware detection have limitations. So, malware detectiontechnique through machine learning is being researched to detect malware variant. However, even in the case of applyingmachine learning, if the proper features of the malware are not properly selected, the machine learning cannot be showncorrectly. We are focusing on feature selection method to find the features of malware variant in this research.

국문 초록 (Abstract)

2018년 시만텍 보고서에 따르면, 모바일 환경에서 변종 악성 앱은 전년도 대비 54% 증가하였고, 매일 24,000개의 악성 앱이 차단되고 있다. 최근 연구에서는 기존 악성 앱 분석 기술의 사용 한계를 파악하고, 신ㆍ변종 악성앱을 탐지하기 위하여 기계학습을 통한 악성 앱 탐지 기법이 연구되고 있다. 하지만, 기계학습을 적용하는 경우에도악성 앱의 특성을 적절하게 선택하여 학습하지 못하면 올바른 결과를 보일 수 없다. 본 연구에서는 신ㆍ변종 악성앱의 특성을 찾아낼 수 있도록 개선된 특성 선택 방법을 적용하여 학습 모델의 정확도를 최고 98%까지 확인할 수있었다. 향후 연구를 통하여 정밀도, 재현율 등 특정 지표의 향상을 목표로 할 수 있다.

참고문헌 (Reference)

• 1 인주호, "효율적인 문서 분류를 위한 혼합 특징 집합과 하이브리드 특징 선택 기법" 한국인터넷정보학회 14 (14): 49-57, 2013
• 2 주진걸, "기계학습 기반의 실시간 악성코드 탐지를 위한 최적 특징 선택 방법" 한국멀티미디어학회 22 (22): 203-209, 2019
• 3 Isabelle Guyon, "introduction to feature extraction" 207 : 1-25, 2006
• 4 Androguard, "androguard"
• 5 R. Veeramani, "Windows API based malware detection and framework analysis" 25-29, 2012
• 6 Yingxian Chang, "Vulnerability Parser: A Static Vulnerability Analysis System for Android Applications" 1288 (1288): 2019
• 7 Symantec, "Symantec internet security threat report" 2018
• 8 L L, "Static analysis of android apps: A systematic literature review" 88 : 67-95, 2017
• 9 Karthik Raman, "Selecting Features to Classify Malware"
• 10 Darell JJ Tan, "Securing android: A survey, taxonomy, and challenges" 47 (47): 2015
• 11 Android Developer, "Safetynet"
• 12 Lei Wang, "Prediction of RNA-Protein Interactions by Combining Deep Convolutional Neural Network with Feature Selection Ensemble Method" 461 : 230-238, 2019
• 13 Zarni Aung, "Permission-Based Android Malware Detection" 2 (2): 228-234, 2013
• 14 M. Ahmadi, "Novel feature extraction, selection and fusion for effective malware family classification" 183-194, 2016
• 15 Scott Schneider, "Modeling goodware characteristics to reduce false positive malware signatures"
• 16 "Mobile threats today"
• 17 Information Security R&D Data Challenge, "Mobile malware"
• 18 Android Developer, "Manifest"
• 19 B. Sun, "Malware family classification method based on static feature extraction" 507-513, 2017
• 20 P. Singhal, "Malware Detection Module Using Machine Learning Algorithms to Assist in Centralized Security in Enterprise Networks" 4 (4): 61-67, 2012
• 21 C. I. Rene, "Malicious Code Intrusion Detection Using Machine Learning And Indicators of Compromise" 15 (15): 160-171, 2017
• 22 Christina Eusanio, "Machine Learning for the Detection of Mobile Malware on Android Devices" Mercyhurst University 19-23, 2019
• 23 K. Rieck, "Learning and classification of malware behavior" 108-125, 2008
• 24 Gartner, "Gartner Says Huawei Secured No. 2 Worldwide Smartphone Vendor Spot, Surpassing Apple in Second Quarter 2018"
• 25 Vipin Kumar, "Feature Selection: A literature Review" 4 (4): 2014
• 26 Fairuz Amalina Narudin, "Evaluation of machine learning classifiers for mobile malware detection" 343-357, 2016
• 27 The Android Open Source Project, "DexDalvik Executable Format"
• 28 M. H. Nguyen, "Auto-Detection of Sophisticated Malware Using Lazy-Binding Control Flow Graph and Deep Learning" 76 : 128-155, 2018
• 29 Keith Makan, "Android Security Cookbook" Packt Publishing 2013
• 30 Isabelle Guyon, "An Introduction to Variable and Feature Selection" 3 : 1157-1182, 2003