Skip to main content
SpringerLink
Log in

Requirements Engineering for Artificial Intelligence: What Is a Requirements Specification for an Artificial Intelligence?

  • Conference paper
  • First Online:
Requirements Engineering: Foundation for Software Quality (REFSQ 2022)

Part of the book series: Lecture Notes in Computer Science ((LNCS,volume 13216))

Abstract

Context: This article concerns requirements for an artificial intelligence (AI) that does a non-algorithmic task that requires real intelligence. Problem: The literature and practice of AI development does not clarify what is a requirements specification (RS) of an AI that allows determining whether an implementation of the AI is correct. Principal ideas: This article shows how (1) measures used to evaluate an AI, (2) criteria for acceptable values of these measures, and (3) information about the AI’s context that inform the criteria and tradeoffs in these measures, collectively constitute an RS of the AI. Contribution: This article shows two related examples of how such an RS can be used and lists some open questions that will be the subject of future work.

This is a preview of subscription content, log in via an institution to check access.

Access this chapter

Log in via an institution
Chapter
USD 29.95
Price excludes VAT (USA)
  • Available as PDF
  • Read on any device
  • Instant download
  • Own it forever
eBook
USD 44.99
Price excludes VAT (USA)
  • Available as EPUB and PDF
  • Read on any device
  • Instant download
  • Own it forever
Softcover Book
USD 59.99
Price excludes VAT (USA)
  • Compact, lightweight edition
  • Dispatched in 3 to 5 business days
  • Free shipping worldwide - see info

Tax calculation will be finalised at checkout

Purchases are for personal use only

Institutional subscriptions

Notes

  1. 1.

    Glossary of Non-Standard Acronyms:

    figure a
  2. 2.

    “E”, “em”, and “er” are gender non-specific third-person singular pronouns in subjective, objective, and possessive forms, respectively.

  3. 3.

    a.k.a.“ML component (MLC)” [16].

  4. 4.

    I.e., there is little certainty on what values of the vague measure are good and are bad. Even when there is certainty that some value is good and another value is bad, there is no certainty about what value in between is the boundary between the good and the bad.

  5. 5.

    It was a total surprise that the cited work was so applicable to RSs for AIs.

References

  1. Ahmad, K., et al.: What’s up with requirements engineering for artificial intelligence systems? In: IEEE 29th RE, pp. 1–12 (2021)

    Google Scholar 

  2. Ashmore, R., et al.: Assuring the machine learning lifecycle: desiderata, methods, and challenges. ACM Comp. Surv. 54(5), 111 (2021)

    Article  Google Scholar 

  3. Berry, D.M.: Empirical evaluation of tools for hairy requirements engineering tasks. Empir. Softw. Eng. 26(6), 1–77 (2021). https://doi.org/10.1007/s10664-021-09986-0

    Article  Google Scholar 

  4. Berry, D.M., et al.: The case for dumb requirements engineering tools. In: REFSQ, pp. 211–217 (2012)

    Google Scholar 

  5. Chuprina, T., et al.: Towards artefact-based requirements engineering for data-centric systems. In: REFSQ-JP 2021: RE4AI (2021)

    Google Scholar 

  6. DiMatteo, J., et al.: Requirements for monitoring inattention of the responsible human in an autonomous vehicle: the recall and precision tradeoff. In: REFSQ-JP 2020: RE4AI (2020)

    Google Scholar 

  7. Hadar, I., Zamansky, A., Berry, D.M.: The inconsistency between theory and practice in managing inconsistency in requirements engineering. Empir. Softw. Eng. 24(6), 3972–4005 (2019). https://doi.org/10.1007/s10664-019-09718-5

    Article  Google Scholar 

  8. Horkoff, J.: Non-functional requirements for machine learning: challenges and new directions. In: IEEE 27th RE, pp. 386–391 (2019)

    Google Scholar 

  9. Hu, B.C., et al.: Towards requirements specification for machine-learned perception based on human performance. In: IEEE 7th AIRE, pp. 48–51 (2020)

    Google Scholar 

  10. Kästner, C.: Machine learning is requirements engineering (2020). https://medium.com/analytics-vidhya/machine-learning-is-requirements-engineering-8957aee55ef4

  11. Kostova, B., et al.: On the interplay between requirements, engineering, and artificial intelligence. In: REFSQ-JP 2020: RE4AI (2020)

    Google Scholar 

  12. Kress-Gazit, H., et al.: Formalizing and guaranteeing human-robot interaction. CACM 64(9), 78–84 (2021)

    Article  Google Scholar 

  13. Mehrabi, N., et al.: A survey on bias and fairness in machine learning. ACM Comp. Surv. 54(6), 1–35 (2021)

    Article  Google Scholar 

  14. Mishra, A.: Metrics to evaluate your machine learning algorithm (2018). https://towardsdatascience.com/metrics-to-evaluate-your-machine-learning-algorithm-f10ba6e38234

  15. Parnas, D.L.: The real risks of artificial intelligence. CACM 60(10), 27–31 (2017)

    Article  Google Scholar 

  16. Rahimi, M., et al.: Toward requirements specification for machine-learned components. In: IEEE 27th RE Workshops (REW), pp. 241–244 (2019)

    Google Scholar 

  17. Salay, R., Czarnecki, K.: Using machine learning safely in automotive software: an assessment and adaption of software process requirements in ISO 26262 (2018). https://arxiv.org/abs/1808.01614

  18. Schmelzer, R.: How do you test AI systems? (2020). https://www.forbes.com/sites/cognitiveworld/2020/01/03/how-do-you-test-ai-systems/

  19. Seshia, S.A., et al.: Towards verified artificial intelligence (2020). https://arxiv.org/abs/1606.08514v4

  20. Valiant, L.: Probably Approximately Correct: Nature’s Algorithms for Learning and Prospering in a Complex World. Basic Books, New York (2013)

    Google Scholar 

  21. Vogelsang, A., Borg, M.: Requirements engineering for machine learning: perspectives from data scientists. In: IEEE 27th RE Workshops (REW), pp. 245–251 (2019)

    Google Scholar 

  22. Zave, P., Jackson, M.: Four dark corners of requirements engineering. TOSEM 6(1), 1–30 (1997)

    Article  Google Scholar 

  23. Zhang, J., Li, J.: Testing and verification of neural-network-based safety-critical control software: a systematic literature review. IST 123, 106296 (2020)

    Google Scholar 

Download references

Acknowledgments

I benefited from voice and text discussions with Krzysztof Czarnecki, Nancy Day, John DiMatteo, Alessio Ferrari, Vijay Ganesh, Andrea Herrmann, Hans-Martin Heyn, Jeff Joyce, Davor Svetinovic, John Thistle, Richard Trefler, and Andreas Vogelsang and his students and post-docs. I thank the anonymous reviewers for their suggestions, only of few of which could be enacted due to the page limit.

Author information

Authors and Affiliations

  1. Cheriton School of Computer Science, University of Waterloo, Waterloo, ON, N2L 3G1, Canada

    Daniel M. Berry

Authors
  1. Daniel M. Berry
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Daniel M. Berry .

Editor information

Editors and Affiliations

  1. University of Pisa, Pisa, Italy

    Vincenzo Gervasi

  2. University of Cologne, Cologne, Germany

    Andreas Vogelsang

Rights and permissions

Reprints and permissions

Copyright information

© 2022 Springer Nature Switzerland AG

About this paper

Check for updates. Verify currency and authenticity via CrossMark

Cite this paper

Berry, D.M. (2022). Requirements Engineering for Artificial Intelligence: What Is a Requirements Specification for an Artificial Intelligence?. In: Gervasi, V., Vogelsang, A. (eds) Requirements Engineering: Foundation for Software Quality. REFSQ 2022. Lecture Notes in Computer Science, vol 13216. Springer, Cham. https://doi.org/10.1007/978-3-030-98464-9_2

Download citation

  • DOI: https://doi.org/10.1007/978-3-030-98464-9_2

  • Published:

  • Publisher Name: Springer, Cham

  • Print ISBN: 978-3-030-98463-2

  • Online ISBN: 978-3-030-98464-9

  • eBook Packages: Computer ScienceComputer Science (R0)

Publish with us

Policies and ethics

Search

Navigation