Published October 21, 2025 | Version v5
Publication Open

Explainability Requirement Practices and Challenges from a Brazilian Industrial Research and Innovation Company

Authors/Creators

  • 1. ROR icon Instituto Federal Goiano
  • 2. ROR icon Universidade Federal de Goiás
  • 3. Universidade Federal do Maranhão
  • 4. Pontifical Catholic University of Rio de Janeiro (PUC-Rio)
  • 5. Federal University of Goiás

Contributors

Contact person:

  • 1. Federal University of Goiás
  • 2. ROR icon Universidade Federal do Maranhão
  • 3. Pontifical Catholic University of Rio de Janeiro (PUC-Rio)
  • 4. ROR icon Universidade Federal de Goiás
  • 5. ROR icon Instituto Federal Goiano

Description

We conducted a study to investigate how professionals leading machine learning (ML) projects perceive explainability and the challenges they face when addressing it as a software requirement. Semi-structured interviews were conducted with 13 professionals responsible for ML projects, following a hypothetico-deductive approach. The collected data were analyzed based on the principles of Grounded Theory, using open and axial coding to identify emerging themes and relationships.

Files

Interview Flow - Explainability in ML Projects.pdf

Files (1.1 MB)

Name Size Download all
md5:c77e8f8eec6804142696e3d34985a7a5
18.1 kB Download
md5:7456b10c90e6e941ffe6a1f4f248aef0
58.4 kB Preview Download
md5:bdc25eb63b52c24d8cae87e589d42605
376.9 kB Preview Download
md5:562c0cc6740983d4e205b04a3d216207
350.1 kB Preview Download
md5:19df1bcf845eab79c2e5c0f8ac586803
270.7 kB Preview Download

Additional details

Related works

Documents
Publication: Semi-Structured Interview (Other)

Dates

Other
2025
Semi-Structured Interview Protocol Document

References

  • Requirements Engineering, Explainability, Machine Learning, Interview