Thesis supervisor: Árpád Beszédes
Location of studies (in Hungarian): University of Szeged Abbreviation of location of studies: SZTE
Description of the research topic:
The traditional software fault localization approaches typically use algorithmic methods to produce the suspicious faulty elements, which are based on the analysis of the source code and associated tests. From the practical point of view, such automatically produced elements are useful only if the real faulty element is located high in this set. Using current approaches, this is hard to achieve, though. Machine learning approaches (for instance, deep neural networks) can achieve high precision in other fields provided they can use a sufficiently large training dataset. The topic of the present proposal is to research on various methods for applying machine learning to software fault localization, and the development of suitable learning datasets, which can significantly enhnace the precision of traditional approaches.
Required language skills: English Further requirements: Knowledge of machine learning approaches is a benefit
Number of students who can be accepted: 2
Deadline for application: 2022-03-15
2024. IV. 17. ODT ülés Az ODT következő ülésére 2024. június 14-én, pénteken 10.00 órakor kerül sor a Semmelweis Egyetem Szenátusi termében (Bp. Üllői út 26. I. emelet).