iSQI CT-AI Exam - Topic 2 Question 13 Discussion

Actual exam question for iSQI's CT-AI exam

Question #: 13
Topic #: 2

An image classification system is being trained for classifying faces of humans. The distribution of the data is 70% ethnicity A and 30% for ethnicities B, C and D. Based ONLY on the above information, which of the following options BEST describes the situation of this image classification system?

SELECT ONE OPTION

AThis is an example of expert system bias.

BThis is an example of sample bias.

CThis is an example of hyperparameter bias.

DThis is an example of algorithmic bias.

Show Suggested Answer

Suggested Answer: A

When written requirements are given in text documents, the best way to generate test cases is by using Natural Language Processing (NLP). Here's why:

Natural Language Processing (NLP): NLP can analyze and understand human language. It can be used to process textual requirements to extract relevant information and generate test cases. This method is efficient in handling large volumes of textual data and identifying key elements necessary for testing.

Why Not Other Options:

Analyzing source code for generating test cases: This is more suitable for white-box testing where the code is available, but it doesn't apply to text-based requirements.

Machine learning on logs of execution: This approach is used for dynamic analysis based on system behavior during execution rather than static textual requirements.

GUI analysis by computer vision: This is used for testing graphical user interfaces and is not applicable to text-based requirements.

by Rickie at Sep 14, 2024, 12:02 PM

Limited Time Offer

25%

Off

Get Premium CT-AI Questions as Interactive Web-Based Practice Test or PDF

Contribute your Thoughts:

Submit Cancel

Kayleigh

3 months ago

70% for one ethnicity is a clear sign of bias.

upvoted 0 times

...

Hector

3 months ago

Not sure if it's just sample bias, could be more complex.

upvoted 0 times

...

Amie

3 months ago

Wait, isn't it also a bit of algorithmic bias?

upvoted 0 times

...

Cyril

4 months ago

I agree, the distribution is skewed.

upvoted 0 times

...

Ardella

4 months ago

This is definitely sample bias.

upvoted 0 times

...

Kathrine

4 months ago

I thought algorithmic bias was more about how the algorithm processes data, not just the data distribution itself. I'm confused!

upvoted 0 times

...

Donte

4 months ago

This question feels familiar; I think we had a similar practice question about data distribution affecting model performance.

upvoted 0 times

...

Rashida

4 months ago

I'm not entirely sure, but I think expert system bias relates more to the decision-making process rather than data distribution.

upvoted 0 times

...

Lashawn

5 months ago

I remember we discussed sample bias in class, especially when the training data is skewed like this. It seems like the right choice.

upvoted 0 times

...

Ora

5 months ago

Okay, I think I've got it. Since the data is heavily biased towards one ethnicity, the model is likely to be biased towards that ethnicity as well. This sounds like a clear case of sample bias to me.

upvoted 0 times

...

Nelida

5 months ago

Hmm, I'm not sure. The question mentions the distribution of the data, but doesn't give any other details about the model or the training process. I'll have to weigh the options carefully before selecting an answer.

upvoted 0 times

...

Renay

5 months ago

I'm a bit confused here. Is this really sample bias, or could it be something else like algorithmic bias? I'll need to think this through carefully.

upvoted 0 times

...

Yuonne

5 months ago

This looks like a classic case of sample bias. The data is heavily skewed towards one ethnicity, which could lead to the model performing poorly on the other ethnicities.

upvoted 0 times

...

Merrilee

5 months ago

Okay, let's see. Policy label and policy signature seem like the two attributes we can fetch using an API. I'm pretty confident about those two choices.

upvoted 0 times

...

Janae

9 months ago

I'm going with option B. This is a textbook case of sample bias. The model is going to be heavily biased towards ethnicity A, and could even completely fail to recognize the other ethnicities.

upvoted 0 times

...

Stephen

9 months ago

Haha, I bet the model will be great at recognizing Smurf faces, but struggle with the rest of us. Classic algorithmic bias!

upvoted 0 times

Jamal

8 months ago

D) This is an example of algorithmic bias.

upvoted 0 times

...

Lon

8 months ago

C) This is an example of hyperparameter bias.

upvoted 0 times

...

Augustine

9 months ago

B) This is an example of sample bias.

upvoted 0 times

...

Mila

9 months ago

A) This is an example of expert system bias.

upvoted 0 times

...

Daniel

9 months ago

I disagree, this looks more like an issue of expert system bias. The developers of the system have likely built in certain assumptions about the prevalence of different ethnicities, leading to this imbalanced dataset.

upvoted 0 times

Rosita

8 months ago

C) I agree, it seems like an example of hyperparameter bias.

upvoted 0 times

...

Darrin

8 months ago

B) I think it's more of a sample bias issue.

upvoted 0 times

...

Ilda

8 months ago

A) This is an example of expert system bias.

upvoted 0 times

...

Gayla

10 months ago

I'm not sure, but I think it could also be D) This is an example of algorithmic bias. The system may not be trained properly to recognize faces from all ethnicities.

upvoted 0 times

...

Dona

11 months ago

I agree with Thaddeus. The data distribution is not representative of the actual population, so it's definitely sample bias.

upvoted 0 times

...

Gaston

11 months ago

This is definitely an example of sample bias. The training data is heavily skewed towards ethnicity A, which could lead to the model performing poorly on the less represented ethnicities.

upvoted 0 times