Training and Validation Datasets Quiz

Quiz covering Core Machine Learning Concepts

Questions in this quiz

A preview of the 5 questions covered. Start the quiz above to answer them, check your score, and read the explanations.

1. 1

   What is the primary purpose of splitting your dataset into training and validation sets?

2. 2

   You are training a classification model and observe that it achieves 99% accuracy on the training data but only 60% accuracy on the validation data. What is the most likely problem?

3. 3

   When performing k-fold cross-validation, what is the role of the validation sets?

4. 4

   Consider a scenario where you have a highly imbalanced dataset for fraud detection (very few fraud cases vs. many non-fraud cases). You split your data into training and validation sets using a standard random split. What potential issue might arise?

5. 5

   You are building a recommendation system. You have a dataset of user-item interactions. You decide to split this data chronologically, using older interactions for training and more recent interactions for validation. What is the primary benefit of this approach over a random split?