Describe Core Machine Learning Concepts AI-900 Practice Questions (Page 7)

Questions 31–33

1. 31application · medium

   A telecom company wants to predict customer churn. They have a dataset with columns: CustomerID, MonthlyCharges, Tenure, ContractType, and Churned (Yes/No). They plan to train a supervised classification model. Which columns should be used as features?

   • AMonthlyCharges, Tenure, Churned
   • BCustomerID, ContractType, Churned
   • CMonthlyCharges, Tenure, ContractType
   • DCustomerID, MonthlyCharges, Tenure
   Reveal Solution Discussion Select an answer first
2. 32application · medium

   A data scientist is preparing a dataset for a binary classification model. They have 10,000 labeled records. To evaluate the model's performance honestly, they split the data into two parts: one part is used to teach the model patterns, and the other part is held out to test how well the model generalizes to new data. Which split is appropriate?

   • AUse 5,000 records for training and 5,000 for validation.
   • BUse all 10,000 records for training and then evaluate on the same data.
   • CUse 2,000 records for training and 8,000 for validation.
   • DUse 8,000 records for training and 2,000 for validation.
   Reveal Solution Discussion Select an answer first
3. 33expert · hard

   A financial firm is building a model to approve small business loans. The dataset includes the applicant's credit score, annual revenue, number of employees, and a historical default flag. The firm wants to deploy the model to make decisions on new applications. Which consideration is most critical when selecting features?

   • AFeatures must be collected after the loan decision
   • BFeatures must be numeric
   • CFeatures must be available at the time of prediction
   • DFeatures must be independent of the label
   Reveal Solution Discussion Select an answer first