Detection of Malware attacks in smart phones using Machine Learning
V. R. Niveditha1, T. V. Ananthan2

1V. R. Niveditha*, Department of Computer Science and Engineering, Dr. M.G.R. Educational and Research Institute, Chennai, India.
2T. V. Ananthan, Department of Computer Science and Engineering, Dr. M.G.R. Educational and Research Institute, Chennai, India.

Manuscript received on October 11, 2019. | Revised Manuscript received on 24 October, 2019. | Manuscript published on November 10, 2019. | PP: 4396-4400 | Volume-9 Issue-1, November 2019. | Retrieval Number: A5082119119/2019©BEIESP | DOI: 10.35940/ijitee.A5082.119119
Open Access | Ethics and Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (

Abstract: In recent years, security has become progressively vital in mobile devices. The biggest security problems in android devices are malware attack which has been exposed to different threats. The volume of new applications by the production of mobile devices and their related app-stores is too big to manually examine the each and every application for malicious behavior. Installing applications which may leads to security vulnerabilities on the smart phones request access to sensitive information. There are various malwares can attack android device namely virus, worms, Botnet, Trojans, Backdoor and Root kits due to these attacks the users is compromised by privacy. Root kits and viruses in mobile phone and IoT devices improve along with smart device versions are very difficult to detect or to the least costly. There are 3 places where the trace of these root kits / virus is visible namely CPU, Baseband and Memory. In the new approach we will use machine learning to detect “anomaly” usage pattern and a remote (master server) will analyze and verify the presence of such threats. This research work aims to develop a pipeline to investigate if any application present in a smart device is a malware or not. This pipeline uses HMM algorithm to read anomaly in application behavior, deep learning with Deep Belief Networks (DBN) to classify application events, and bootstrapping algorithm using random forest to categorize the application itself after malware or benign.
Keywords: Malware Attacks, Mobile Phone, Android Device, Hidden Markov Model (HMM), Deep Belief Network (DBN), Random Forest (RF), Machine Learning (ML)
Scope of the Article: Machine Learning