Discuss Isaca AAIA Exam Topic 1 Question 22

Which of the following pre-processing steps would MOST effectively justify an AI model's decision to a non-technical stakeholder?

APermutation feature importance

BLocal interpretable model-agnostic explanations (LIME)

CPartial dependence plots

DAI model penetration testing

Suggested Answer: B

While all the listed techniques (except penetration testing) support interpretability, 'LIME' is specifically noted in the ISACA AAIA Study Guide for its ability to explain individual decisions. LIME creates a simpler, interpretable model around a specific prediction to show which features (e.g., high income or low debt) were the primary drivers for that specific case. This 'Local' explanation is much easier for non-technical stakeholders or customers to understand than 'Global' metrics like feature importance (Option A) or partial dependence plots (Option C), which describe the model's behavior as a whole.

Get Premium AAIA Questions as Interactive Web-Based Practice Test or PDF Real Exam Format Detailed Explanations Instant Access Get Your Practice Test

12 comments · newest first

Herminia 1 month ago

A) Permutation feature importance is good too, but less intuitive.

Reena 1 month ago

I agree, LIME is user-friendly for non-techies.

Tish 1 month ago

I think B) LIME is the best choice. It explains predictions clearly.

Mohammad 2 months ago

I’m not convinced LIME is always effective for non-techies.

Rashad 2 months ago

A) Permutation feature importance is also useful, but not as straightforward.

Devora 2 months ago

Wait, isn't D) AI model penetration testing more about security than explanations?

Iola 2 months ago

Totally agree, LIME makes it super clear!

Glory 2 months ago

I think B) LIME is the best choice for explaining AI decisions.

Lon 2 months ago

I don't recall much about AI model penetration testing in this context; it seems more about security than explaining decisions.

Brett 3 months ago

I feel like partial dependence plots could also be useful, but they might be too technical for some stakeholders.

Brock 3 months ago

I'm not entirely sure, but I remember permutation feature importance being mentioned in class as a way to show how features impact predictions.

Rosalyn 3 months ago

I think LIME might be the best choice since it provides local explanations that are easier for non-technical people to understand.

Isaca AAIA Exam - Topic 1 Question 22 Discussion