An Enhanced Method for Identifying Android Malware Detection
P. Jayanthi1, K. Nirmaladevi2, N. Krishnamoorthy3
1Dr. Jayanthi P.Department of Computer Science and Engineering, Kongu Engineering College, Perundurai, Erode, (Tamil Nadu), India.
2Dr.Nirmaladevi K., Department of Computer Science and Engineering, Kongu Engineering College, Perundurai, Erode, (Tamil Nadu), India.
3Dr. Krishnamoorthy N., Department of Computer Science and Engineering, Kongu Engineering College, Perundurai, Erode, (Tamil Nadu), India.
Manuscript received on November 17., 2019. | Revised Manuscript received on November 24 2019. | Manuscript published on 30 November, 2019. | PP: 12871-12875 | Volume-8 Issue-4, November 2019. | Retrieval Number: D5307118419/2019©BEIESP | DOI: 10.35940/ijrte.D5307.118419
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 (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: In worldwide, all people are living with mobile applications for most of their life time. A statistical survey shows that mobile user exceeds 5 billions by 2019. There is a necessity to download different kinds of applications in different occasions. The library in android OS used for displaying media content has multiple vulnerabilities which enable the attackers to exploit media files and run the malicious code. The new ranges of attacks have been opened up today. The malware application does fraudulent activities automatically in the mobile without the knowledge of users. It is very difficult to identify the malware among such applications. Thus a challenge rises for protecting the mobile phones from these attacks. The existing method, “Significant Permission Identification for Machine-Learning-Based Android Malware Detection (SIGPID)”, which uses Multi-Level Data Pruning process to identify significant permissions. In SIGPID, three level pruning process namely Permission Ranking with Negative Rate (PRNR), Support based Permission Ranking (SPR) and Permission Mining with Association Rule (PMAR) are applied to the dataset followed by SVM classification. The large dimension of the dataset negatively affects the malware detection efficiency. To reduce features of malicious apps further, an enhanced method called “Enhanced Model of Significant Permission Identification (ESID)” is proposed to identify android malware applications using data mining techniques. It adds the process to remove non-significant permissions and to classify the benign apps and malicious apps using SVM before installing an android application in the mobile. The experimental result shows that the better accuracy of 93.75% in identifying the malicious apps..
Keywords: Android Malware Detection, Datamining Techniques, Rank Based Approach, Feature Reduction.
Scope of the Article: Emulation and Simulation Methodologies for IoT.