CertNexus Exam AIP-210 Topic 5 Question 30 Discussion

Actual exam question for CertNexus's AIP-210 exam

Question #: 30
Topic #: 5

You create a prediction model with 96% accuracy. While the model's true positive rate (TPR) is performing well at 99%, the true negative rate (TNR) is only 50%. Your supervisor tells you that the TNR needs to be higher, even if it decreases the TPR. Upon further inspection, you notice that the vast majority of your data is truly positive.

What method could help address your issue?

ANormalization

BOversampling

CPrincipal components analysis

DQuality filtering

Show Suggested Answer

Suggested Answer: B

Oversampling is a method that can help address the issue of imbalanced data, which is when one class is much more frequent than the other in the dataset. This can cause the model to be biased towards the majority class and have a low true negative rate. Oversampling involves creating synthetic samples of the minority class or replicating existing samples to balance the class distribution. This can help the model learn more from the minority class and improve the true negative rate. Reference: [Handling imbalanced datasets in machine learning], [Oversampling and undersampling in data analysis - Wikipedia]

by Dean at Sep 14, 2024, 09:22 PM

Limited Time Offer

25%

Off

Get Premium AIP-210 Questions as Interactive Web-Based Practice Test or PDF

Contribute your Thoughts:

Submit Cancel

Shelia

2 days ago

I think normalization is more about scaling, not class balance.

upvoted 0 times

...

Tamie

8 days ago

Oversampling could help balance the classes!

upvoted 0 times

...

Theola

14 days ago

Quality filtering sounds familiar, but I can't recall if it directly relates to improving TNR in this context.

upvoted 0 times

...

Lashon

19 days ago

I think we practiced a question where we had to improve TNR, and oversampling was one of the suggested methods.

upvoted 0 times

...

Janella

24 days ago

I'm not entirely sure, but I think normalization is more about scaling features rather than addressing class imbalance.

upvoted 0 times

...

Gail

1 month ago

I remember discussing how oversampling can help balance classes in a dataset, especially when one class is much larger than the other.

upvoted 0 times

...

Corrie

1 month ago

Okay, I think I've got a handle on this. Since the supervisor wants to increase the TNR, even if it means a slight decrease in TPR, oversampling the negative class seems like the best option. I'll make sure to try that out and see how it affects the model's performance.

upvoted 0 times

...

Stephaine

1 month ago

I'm a bit confused on this one. The question mentions that the model has high accuracy, so I'm not sure if normalization or PCA would be the best approach. Maybe quality filtering could help, but I'm not sure how that would impact the TPR and TNR specifically.

upvoted 0 times

...

Catarina

1 month ago

Oversampling the negative class definitely seems like the way to go here. With the data being heavily skewed towards the positive class, that's likely the root cause of the low TNR. I'll make sure to try that out.

upvoted 0 times

...

Renea

1 month ago

Hmm, this seems like a tricky one. I'm not sure if normalization would help here since the issue seems to be with the class imbalance. Maybe oversampling the negative class could be a good approach to boost the TNR.

upvoted 0 times

...