Bibliograph. Daten | Buchholz, Max: Deep learning in streamlining the conversion of natural language requirements into template-based architecture. Universität Stuttgart, Fakultät Informatik, Elektrotechnik und Informationstechnik, Bachelorarbeit Nr. 90 (2023). 75 Seiten, englisch.
|
Kurzfassung | This thesis evaluates how recent development in Machine Learning, especially Large Language Models could be used to help with the conversion of natural language based requirement statements into a template-based structure. To evaluate different approaches, they have been implemented, tested and evaluated, on an existing dataset, complemented by converted requirement statements. It seems that the tested approaches are feasible to be used and further investigated, as they were already able to provide a decent performance. Therefore, machine learning based conversion of requirement statements can improve the process of converting existing statements into a normed structure.
|
Volltext und andere Links | Volltext
|
Abteilung(en) | Universität Stuttgart, Institut für Softwaretechnologie, Empirisches Software Engineering
|
Betreuer | Wagner, Prof. Stefan; Habib, Mohammad Kasra |
Eingabedatum | 5. April 2024 |
---|