Tobias Hey

Postdoc at Karlsruhe Institute of Technology (KIT)

tobias_hey.jpg

I am researching in the domain of Natural Language Processing (NLP) for Software Engineering (SE) with focus on traceability link recovery and requirements engineering.

In my dissertation, I developed the automated traceability link recovery approach FTLR, that is able to relate requirements to their corresponding source code entities by utilizing fine-grained word embedding-based relations. Furthermore, I’ve developed the requirements classification approach NoRBERT and integrated its results as a filter into FTLR.

Previously, I’ve done research on programming in natural language, mapping natural language instructions to their corresponding API calls.

selected publications

  1. RE
    Requirements Classification for Traceability Link Recovery
    Tobias HeyJan Keim, and Sophie Corallo
    In 2024 IEEE 32nd International Requirements Engineering Conference (RE), 2024
  2. ICSE
    Recovering Trace Links Between Software Documentation And Code
    Jan KeimSophie CoralloDominik FuchßTobias Hey, Tobias Telge, and Anne Koziolek
    In 2024 IEEE/ACM 46th International Conference on Software Engineering (ICSE), Apr 2024
  3. ICSME
    Improving Traceability Link Recovery Using Fine-grained Requirements-to-Code Relations
    Tobias Hey, Fei Chen, Sebastian Weigelt, and Walter F. Tichy
    In 2021 IEEE International Conference on Software Maintenance and Evolution (ICSME), Apr 2024
  4. RE
    NoRBERT: Transfer Learning for Requirements Classification
    Tobias HeyJan KeimAnne Koziolek, and Walter F. Tichy
    In 2020 IEEE 28th International Requirements Engineering Conference (RE), Apr 2024
  5. ACL
    Programming in Natural Language with fuSE: Synthesizing Methods from Spoken Utterances Using Deep Natural Language Understanding
    Sebastian Weigelt, Vanessa Steurer, Tobias Hey, and Walter F. Tichy
    In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, Apr 2024