This topic is concerned to the application of Artificial Intelligence for enhancing Software Engineering activities, particularly Software Testing (ST) activities. There is a large scope of tasks in ST that can be improved by the use of AT techniques like machine learning and meta-heuristics. Machine learning can be adopted to predicting software actifacts that are prone to defects (the Bug Predicition problem) in such a way to prioritize ST effort to more complex artifacts. Test Case Selection can be performed using meta-heuristics (e.g., genetic algorithms, swarm intelligence) to optimize evaluation metrics like code or requirements coverage. Finally once bugs are found in a SW, bug reports are opened and stored in issue tracking systems. Bug repositories can contain a large and unstructured corpus of textual bug reports, which is the scope of Bug Report Mining tasks, like detecting duplicated reports, classification of difficulty, among others.
A Bispo, R Prudêncio, D Véras. Instance Selection and Class Balancing Techniques for Cross Project Defect Prediction. 2018 7th Brazilian Conference on Intelligent Systems (BRACIS), 552-557
LS De Souza, RBC Prudêncio, FA Barros, EHS Aranha. Search based constrained test case selection using execution effort. Expert systems with applications 40 (12), 4887-4896
M Takaki, D Cavalcanti, R Gheyi, J Iyoda, M d’Amorim, RBC Prudêncio. Randomized constraint solvers: a comparative study. Innovations in Systems and Software Engineering 6 (3), 243-253
I Santos, J Araújo, C Lima, RBC Prudêncio, F Barros. AVS: An approach to identifying and mitigating duplicate bug reports. Proceedings of the XIV Brazilian Symposium on Information Systems, 22