MH-DLdroid: A Meta-Heuristic and Deep Learning-Based Hybrid Approach for Android Malware Detection
Ravi Sharma, Chaitanya Agrawal, H Gao, S Cheng, W Zhang, R Feng, S Chen, X Xie, G Meng, S Lin, Y Liu, M Cai, Y Jiang, C Gao, H Li, W Yuan, S Sasidharan, C Thomas, S Millar, N Mclaughlin, J Rincon, P Miller, T Lu, Y Du, L Ouyang, Q Chen, X Wan, N Zhang, Y Tan, C Yang, Y Zhangli, A Mahindru, A Sangal, T Kim, B Kang, M Rho, S Seze, E Im, Y Yang, X Du, Z Yang, X Liu, Yang, X Xiao, S Zhang, F Mercaldo, G Hu, A Sangaiah, Y Hei, R Yang, H Peng, L Wan, X Xu, J Liu, H Liu, J Xu, L Sun, F Ou, J Xu, W Zhang, N Luktarhan, C Ding, B Lu, T Frenklach, D Cohen, A Shabtai, R Puzis, L Vu, S Jung, P Xu, C Eckert, A Zarras, M Norouzian, P Xu, C Eckert, A Zarras, A Mahindru, A Sangal, R Surendran, T Thomas, S Emmanuel, X Jiang, B Mao, J Guan, Huang, H Bai, N Xie, X Di, Q Ye, R Taheri, M Ghahramani, R Javidan, M Shojafar, Z Pooranian, M Contic, Y Ding, X Zhang, J Hu, W Xu, A Arora
Abstract
With the fast advancement of smartphone technology, the smartphone has emerged as the most prevailing instrument for accessing the Internet and obtaining a wide range of services with a click. Increased use of smartphones for online payments has attracted fraudsters, adding to an increase in malware outbreaks. Mobile application vulnerabilities and malware are the origins of various types of fraud and numerous cyber-attacks. Large datasets are frequently used for malware analysis; however, large datasets may contain many redundant, inappropriate, and noisy features, resulting in misclassification and low detection rates. This paper presents a hybrid approach to Android malware detection that reduces the dimensionality of datasets to reduce resource-intensive computation while preserving critical information. We present a novel hybrid approach for detecting Android malware based on a metaheuristic (modified Intelligent Water Drop Algorithm (IWD)) and Deep Learning (DL) techniques. The studies show that the proposed approach efficiently removes irrelevant attributes and attains significant detection performance with an F1-Score of 93.7%, a precision of 95.35, an accuracy of 99.12%, and a recall rate of 96.68%.