Permission Sensitivity-Based Malicious Application Detection for Android

Since a growing number of malicious applications attempt to steal users’ private data by illegally invoking permissions, application stores have carried out many malware detection methods based on permissions. However, most them ignore specific permission combinations and categories that affect the accuracy. The features they extracted are neither representative enough distinguish benign applications. For these problems, an Android method sensitivity is proposed. First, for each kind categories, combination extracted. sensitive feature set corresponding category label then obtained selection sensitivity. In following step, call situation be detected compared with set, weight allocation used quantify this information into numerical features. proposed detection, three machine-learning algorithms selected construct classifier model optimize parameters. Compared traditional methods, consumed 60.94% less time while still achieving high accuracy up 92.17%.