Evaluation of Network Traffic Analysis Using Fuzzy C-Means Clustering Algorithm in Mobile Malware Detection

Due to widespread use of mobile devices and open source nature of Android operating system, such devices have been targeted by attackers. The Android malware steadily grow in number and complexity. This motivates researchers to develop detection methods. In this paper, we introduce the use of Fuzzy C-Means clustering in Android malware detection. We chose 800 malware samples and 100 clean applications, and collected generated network traffic. Selected features were extracted from the network traffic, and then used in Fuzzy C-Means clustering algorithm. The results show that this algorithm is capable of clustering our data into two groups of clean and malicious data. Furthermore, we validated our results by comparing them to our labelled dataset, which showed 13% discrepancy in results.