Session-Based Recommender Systems for Action Selection in GUI Test Generation
This work addresses the challenge of creating more effective and human-like test sequences for GUI testing, which is incremental as it builds on existing techniques.
The paper tackled the problem of action selection in GUI test generation by proposing a session-based recommender system to mimic past user behavior, achieving preliminary results that show the approach can significantly improve GUI-based test generation.
Test generation at the graphical user interface (GUI) level has proven to be an effective method to reveal faults. When doing so, a test generator has to repeatably decide what action to execute given the current state of the system under test (SUT). This problem of action selection usually involves random choice, which is often referred to as monkey testing. Some approaches leverage other techniques to improve the overall effectiveness, but only a few try to create human-like actions---or even entire action sequences. We have built a novel session-based recommender system that can guide test generation. This allows us to mimic past user behavior, reaching states that require complex interactions. We present preliminary results from an empirical study, where we use GitHub as the SUT. These results show that recommender systems appear to be well-suited for action selection, and that the approach can significantly contribute to the improvement of GUI-based test generation.