Microsoft Exam AI-900 Topic 2 Question 75 Discussion

Actual exam question for Microsoft's AI-900 exam

Question #: 75
Topic #: 2

[All AI-900 Questions]

What should you do to reduce the number of false positives produced by a machine learning classification model?

AInclude test data in the training data.

BIncrease the number of training iterations.

CModify the threshold value in favor of false positives.

DModify the threshold value in favor of false negatives.

Show Suggested Answer

Suggested Answer: A

by Rory at Sep 18, 2024, 09:56 AM

Limited Time Offer

25%

Off

Get Premium AI-900 Questions as Interactive Web-Based Practice Test or PDF

Contribute your Thoughts:

Submit Cancel

Benedict

3 months ago

D all the way, baby! Embracing those false negatives is the key to a happy and healthy classification model.

upvoted 0 times

Lilli

2 months ago

D) Modify the threshold value in favor of false negatives.

upvoted 0 times

...

Johana

2 months ago

C) Modify the threshold value in favor of false positives.

upvoted 0 times

...

Tamar

2 months ago

B) Increase the number of training iterations.

upvoted 0 times

...

Camellia

2 months ago

A) Include test data in the training data.

upvoted 0 times

...

Janine

3 months ago

I'm gonna have to go with C on this one. Gotta love those false positives, they're like surprises in a box of machine learning treats!

upvoted 0 times

...

Eleonore

3 months ago

Wait, isn't the answer A? Including the test data in the training data seems like a no-brainer to me. What could go wrong?

upvoted 0 times

...

Joni

3 months ago

Aha! C is the answer. Modifying the threshold to favor false positives will definitely help reduce the number of false positives.

upvoted 0 times

Nguyet

2 months ago

Including test data in the training data can also help improve the model's performance.

upvoted 0 times

...

Merlyn

3 months ago

That makes sense. It's important to balance between false positives and false negatives in a classification model.

upvoted 0 times

...

Fernanda

3 months ago

I think C is the answer. Modifying the threshold value can help reduce false positives.

upvoted 0 times

...

Janna

4 months ago

Hmm, I think D is the way to go. Modifying the threshold to favor false negatives sounds like a good strategy to me.

upvoted 0 times

Craig

3 months ago

Interesting perspective! I think B could also be helpful. Increasing the number of training iterations might improve the model's accuracy overall.

upvoted 0 times

...

Millie

3 months ago

Actually, I think C might be a good compromise. Modifying the threshold value in favor of false positives could help balance the model's performance.

upvoted 0 times

...

Yuonne

3 months ago

I see your point, but I still think D is the better choice. Modifying the threshold for false negatives can be more beneficial in some cases.

upvoted 0 times

...

Rana

3 months ago

I disagree, I think A is the best option. Including test data in the training data can help reduce false positives.

upvoted 0 times

...

Casie

4 months ago

Yes, but it's a trade-off. We need to find the right balance to reduce false positives without increasing false negatives too much.

upvoted 0 times

...

Virgina

4 months ago

I'm pretty sure the answer is B. Increasing the number of training iterations should help the model learn better and reduce false positives.

upvoted 0 times

Kyoko

4 months ago

I agree with you, increasing the number of training iterations can also help reduce false positives.

upvoted 0 times

...

Nu

4 months ago

I think the answer is C. Modifying the threshold value in favor of false positives should help reduce them.

upvoted 0 times

...

Gearldine

4 months ago

But wouldn't that increase the number of false positives?

upvoted 0 times

...

Casie

5 months ago

I think we should modify the threshold value in favor of false positives.

upvoted 0 times

...