GUI Software Fault Localization Using N-gram Analysis

Graphical User Interfaces (GUIs) have become an important and accepted way of interacting with today's software. Fault localization is considered to be one of the most expensive program debugging activities. This paper presents a fault localization technique designed for GUI software. Unlike traditional software, GUI test cases usually are event sequences and each individual event has a unique corresponding event handler. We apply data mining techniques to the event sequences and their output data in terms of failure detections collected in the testing phase to rank the fault proneness of the event handlers for fault localization. Our method applies N-gram analysis to rank the event handlers of GUI programs and data collected from case studies on four real life GUI programs demonstrate the effectiveness of the proposed technique.