Describe Core Machine Learning Concepts AI-900 Practice Questions (Page 1)

Questions 1–5

1. 1foundation · easy

   Which of the following best describes a label in a machine learning dataset?

   • AA metric used to evaluate model performance
   • BAn input variable that the model uses to make predictions
   • CA portion of the dataset set aside to test the model
   • DThe known outcome or target value that the model learns to predict
   Reveal Solution Discussion Select an answer first
2. 2application · medium

   A manufacturing company is training a model to detect defective parts from sensor readings. The team splits the labeled dataset into two parts: 80% for training and 20% for validation. After training, the model achieves 99% accuracy on the training set but only 82% accuracy on the validation set. What does this difference most likely indicate?

   • AThe model has not been trained long enough
   • BThe validation set is larger than the training set
   • CThe model is overfitting the training data
   • DThe validation set contains only defective parts
   Reveal Solution Discussion Select an answer first
4. 3application · medium

   A manufacturing company wants to predict whether a machine will fail within the next week. They have historical sensor data with columns: Temperature, Vibration, Pressure, and Failure (Yes/No). What should be the label in this supervised learning scenario?

   • AThe Failure column (Yes/No)
   • BThe Vibration column
   • CThe Pressure column
   • DThe Temperature column
   Reveal Solution Discussion Select an answer first
5. 4expert · hard

   A data science team is building a churn prediction model. They have a dataset with 100,000 customers, but only 2% churned. They split the data into 80% training and 20% validation. The model achieves 98% accuracy on the validation set, yet the business team reports the model is not useful because it predicts 'no churn' for almost every customer. What is the most likely explanation?

   • AAccuracy is misleading for imbalanced data, and the model is not identifying churners
   • BThe validation set was too small to contain enough churn examples
   • CThe training set was too large, causing the model to overfit
   • DThe validation set was not split chronologically
   Reveal Solution Discussion Select an answer first
6. 5application · medium

   A retail company wants to build a machine learning model to predict whether a customer will make a purchase in the next 30 days. They have historical data with columns: CustomerID, Age, LastPurchaseDate, TotalSpendLastYear, and PurchasedNext30Days. The company plans to train a supervised classification model. Which columns should be used as features and which as the label?

   • AFeatures: Age, LastPurchaseDate, TotalSpendLastYear; Label: PurchasedNext30Days
   • BFeatures: PurchasedNext30Days, Age, TotalSpendLastYear; Label: LastPurchaseDate
   • CFeatures: CustomerID, LastPurchaseDate, TotalSpendLastYear; Label: Age
   • DFeatures: CustomerID, Age, LastPurchaseDate; Label: TotalSpendLastYear
   Reveal Solution Discussion Select an answer first