Practice Imbalanced Datasets (12.4.D) - Model Evaluation and Validation

Practice: Imbalanced Datasets

Lesson 12.4.DModel Evaluation and Validation

Practise this section

9 questions

Flashcard drill
4 cards from this section. Quick, good the night before a test.
Exercise
6 questions · about 4 min · pass mark 70%. Hints available.
Sign up to start the exercise
Quiz
3 questions · about 2 min · pass mark 70%. Hints available.
Sign up to start the quiz

6 sample questions

Question 1.
What is an imbalanced dataset?
Show hintHide hint
Think about how many instances you have of each class.
Question 2.
Name one metric used for evaluating imbalanced datasets.
Show hintHide hint
Consider metrics that focus on positives.
3Question 3.
What is a key metric in evaluating imbalanced datasets?
- Accuracy
- Recall
- F1-Score
Show hintHide hint
Think about metrics that account for both false positives and negatives.
4Question 4.
True or False: Accuracy is always a reliable metric for model evaluation.
- True
- False
Show hintHide hint
Consider situations where one class vastly outnumbers another.
Question 5.
Given a dataset with two classes, A (90% samples) and B (10% samples), how would you evaluate a model trained on this data? What metrics would you focus on and why?
Show hintHide hint
Consider the implications of performance metrics when classes are imbalanced.
Question 6.
Explain how you would implement SMOTE in a practical case and discuss its potential pitfalls.
Show hintHide hint
Think about balancing quantities and avoiding mimicking too closely.

5 more questions in this section. Enroll free to practise them all.