An Innovative Technique to Detect Malicious Applications in Android
Journal: International Journal of Science and Research (IJSR) (Vol.4, No. 12)Publication Date: 2015-12-05
Authors : Sharvari Prakash Chorghe; Narendra Shekokar;
Page : 1846-1849
Keywords : Android malware detection; improved feature selection; bayesian classification; data mining; machine learning;
Abstract
Android Operating System is getting popular with majority of users of the entire genre. Ever increasing mobile malware is a major threat to Android OS. Hence it is essential to add an extra security layer to android smart phone. Various malware detection techniques use Mutual Information or Information Gain for feature selection. Mutual Information calculates relevance of a feature and the class variable. Feature selection or ranking is crucial step before applying any machine learning classification algorithm on the pre-processed data. The proposed system uses improved mutual information feature selection (IMIFS) method to maximize relevance and minimize redundancy in the selected features. This improves the accuracy of the classification model. The proposed system will use features from manifest. xml as well as application code for malware detection. Bayesian Classification is used to develop classifier model to detect malicious applications before installation on the device.
Other Latest Articles
- Transition from IPv4 to IPv6 and their Security Challenges
- A Review of Enhanced Technique for Cloud Data Storage Security Using Data Slicing
- A Study of Malaria Epidemic Model Using the Effect of Lost Immunity
- Combination of Fingerprints to Secure Privacy
- Minimum Spanning Tree Determination Program Using Kruskal Algorithm on Visual Basic 6.0
Last modified: 2021-07-01 14:28:06