The increasing number of malwares has led to an increase in research work on malware analysis studying the malware behavior. The malware tries to leak sensitive information from infected devices. In this paper, we study a specific attack method, which distributes the data source and the point of data loss on different versions of the malware application. That is done using local storage by storing part or all of the vital data to be leaked in the future.
We introduce a Distributed Malware Detection Algorithm (DMDA), which is an algorithm to detect distributed malware on app versions. DMDA proposes a new way to analyze application against redistributed malware. DMDA is created to analyze the data and identify transitional loss points. We test this algorithm on a sample of Android applications published on the Google Play market containing 100 applications, where each application has two versions. The algorithm detected 150 transient data sources, 200 transient loss of data point and two leakages of data. In comparison, this dataset was checked using 56 anti-malware applications but none of them could find any malicious code.