Contrasting Permission Patterns between Clean and Malicious Android Applications

摘要: The Android platform uses a permission system model to allow users and developers regulate access private information resources required by applications. Permissions have been proved be useful for inferring behaviors characteristics of an application. In this paper, novel method extract contrasting patterns clean malicious applications is proposed. Contrary existing work, both used permissions were considered when discovering the patterns. We evaluated our methodology on malware dataset, each comprising 1227 Our empirical results suggest that can capture key differences between applications, which assist in characterizing these two types