MACHINE LEARNING APPROACH TO DETECT ANDROID MALWARESJournal: International Journal of Engineering Sciences & Research Technology (IJESRT) (Vol.5, No. 3)
Publication Date: 2016-03-30
Authors : Nikita Rai; Dr Tri p ti Arjariya;
Page : 452-458
Keywords : M alicious; A ndroid; C lassification; Naïve Bayes;
Mobile phone industry is growing at rapid speed .These mobile ph o nes are running on diiferent platform suc h as JAVA, Android, IOS, S ysmbian and others. Out all these platforms An dr oid c over maximum share amount Smart phone platform. And r oid platforms supports millions of applications that can be download from v arious repositories such as g oogle play. These applications are installed and used. The applications present in these repository may be malicious which leady to security problems using these application. In this paper an effective approach has been proposed for detect ion of the malicious application based on the permission gro u ps. In proposed work, binary classification of application s are carried out into two label i.e. B enign and malicious one . In this de veloped approach t he distinguished features are evaluated and filtered out using features evaluation technique such as Information gain, Gain ratio, Gini Index, Chi - square test. Finally based on the features evaluated the classification is done using super vised machine learning techniques.
Other Latest Articles
Last modified: 2016-03-16 18:17:02