Exam DP-100 topic 5 question 21 discussion

Actual exam question from Microsoft's DP-100

Question #: 21
Topic #: 5

You are building a binary classification model by using a supplied training set.
The training set is imbalanced between two classes.
You need to resolve the data imbalance.
What are three possible ways to achieve this goal? Each correct answer presents a complete solution.
NOTE: Each correct selection is worth one point.

A. Penalize the classification
B. Resample the dataset using undersampling or oversampling
C. Normalize the training feature set
D. Generate synthetic samples in the minority class
E. Use accuracy as the evaluation metric of the model

Show Suggested Answer

Suggested Answer: ABD 🗳️

by ljljljlj at July 11, 2021, 2:25 p.m.

Comments

Submit Cancel

ljljljlj

Highly Voted 3 years, 6 months ago

On exam 2021/7/10

upvoted 6 times

...

Matt2000

Most Recent 11 months, 1 week ago

This reference might be useful: https://machinelearningmastery.com/tactics-to-combat-imbalanced-classes-in-your-machine-learning-dataset/

upvoted 1 times

...

michaelmorar

2 years ago

Selected Answer: ABD

using accuracy as a metric hardly answers the question, and normalising the feature set won't help the class imbalance.

upvoted 2 times

...

AP_15

2 years, 1 month ago

Selected Answer: ABD

correct

upvoted 2 times

...

synapse

2 years, 10 months ago

Selected Answer: ABD

Penalize model Use over/under sampling SMOTE. There's another question of similar type which ignore Penalizing Sample as an option. This answer is correct.

upvoted 4 times

...

dijaa

3 years, 4 months ago

correct!

upvoted 3 times

...