Multiple Concurrency Anomalies Classification for Mobile Applications using Support Vector Machine

Abstract

Mobile applications are integral part of daily life due to their portability and convenience. In recent research, mobile applications are facing the uniqueness of approach for specific anomaly and large number of false positive. In this study, we propose Support Vector Machine (SVM) based concurrency anomaly classification approach to dynamically distinguish the status in runtime. By using anomaly classification, the approach is enabled to classify multiple concurrency anomaly with vector clocks. We proposed anomalies classification for mobile applications to detect the potential exception and reduce the false positive in runtime.