Practice Exam Questions

---

Question 1

You are training a model to predict house prices. The dataset includes columns for square footage, number of bedrooms, location, and sale price.
Which column is the label?

A. Square footage
B. Number of bedrooms
C. Location
D. Sale price

Correct Answer: D

Explanation:
The label is the value the model is trained to predict. In this scenario, the goal is to predict the sale price.

---

Question 2

Which statement best describes a feature in a machine learning dataset?

A. The final prediction made by the model
B. An input value used to make predictions
C. A rule written by a developer
D. The accuracy of the model

Correct Answer: B

Explanation:
Features are the input variables that provide information the model uses to make predictions.

---

Question 3

A dataset contains customer age, subscription length, monthly charges, and whether the customer canceled the service.
What is the label?

A. Customer age
B. Subscription length
C. Monthly charges
D. Whether the customer canceled

Correct Answer: D

Explanation:
The label represents the outcome being predicted—in this case, whether the customer canceled the service.

---

Question 4

Which type of machine learning requires both features and labels?

A. Unsupervised learning
B. Reinforcement learning
C. Supervised learning
D. Clustering

Correct Answer: C

Explanation:
Supervised learning uses labeled data so the model can learn the relationship between features and known outcomes.

---

Question 5

A dataset is used to group customers based on purchasing behavior, but it does not contain any target outcome.
What does this dataset contain?

A. Labels only
B. Features only
C. Training results
D. Predictions

Correct Answer: B

Explanation:
Unsupervised learning datasets contain features but do not include labels.

---

Question 6

In an email spam detection dataset, which item would most likely be a feature?

A. Spam or not spam
B. Model accuracy score
C. Number of words in the email
D. Final prediction

Correct Answer: C

Explanation:
The number of words is an input characteristic used by the model to make predictions, making it a feature.

---

Question 7

Which statement about labels is TRUE?

A. Labels are optional in supervised learning
B. Labels are the inputs used by the model
C. Labels represent the value the model predicts
D. Labels are created after predictions are made

Correct Answer: C

Explanation:
Labels are the known outcomes the model is trained to predict in supervised learning scenarios.

---

Question 8

You are preparing data in Azure Machine Learning to predict product demand.
Which columns should be selected as features?

A. Only the column you want to predict
B. All columns except the target outcome
C. Only numerical columns
D. Only categorical columns

Correct Answer: B

Explanation:
Features are the input columns used to predict the target outcome, which is the label.

---

Question 9

A dataset includes the following columns: temperature, humidity, wind speed, and weather condition.
If the goal is to predict the weather condition, what are temperature, humidity, and wind speed?

A. Labels
B. Predictions
C. Features
D. Outputs

Correct Answer: C

Explanation:
These values are inputs used to predict the weather condition, making them features.

---

Question 10

Which scenario best represents a labeled dataset?

A. Customer data grouped by similarity
B. Sensor readings without outcomes
C. Product reviews with sentiment categories
D. Website logs without classifications

Correct Answer: C

Explanation:
Product reviews with sentiment categories include known outcomes, which are labels, making the dataset labeled.

---

Exam Pattern Tip

On AI-900:

• Features = inputs
• Labels = outputs
• If labels exist → supervised learning
• If no labels → unsupervised learning

If you can identify those quickly, you’ll eliminate most wrong answers immediately.

---

Go to the AI-900 Exam Prep Hub main page.