Supervised Learning CT-AI Practice Questions (Page 6)

Questions 26–27

1. 26application · medium

   A data scientist trains a model on a dataset and achieves 99% accuracy on the training data but only 70% accuracy on a held-out test set. What is the most likely issue, and what is the best next step?

   • AUnderfitting; the model needs more features
   • BOverfitting; the model should be simplified or regularized
   • CThe test set is too small; more data should be collected
   • DData leakage; the test set should be re-split
   Reveal Solution Discussion Select an answer first
2. 27foundation · easy

   A team wants to build a model that predicts whether a customer will churn based on historical customer data. The dataset includes a column 'churned' with values 'yes' or 'no' for each customer. Which ML paradigm is most appropriate?

   • AUnsupervised learning
   • BSemi-supervised learning
   • CReinforcement learning
   • DSupervised learning
   Reveal Solution Discussion Select an answer first