Microsoft Azure AI Fundamentals (AI-900) Practice Exam

What type of machine learning task is predicting overtime hours based on the number of orders received?

1. Classification

2. Regression

3. Clustering

4. Anomaly Detection

The correct answer is: Regression

Predicting overtime hours based on the number of orders received is an example of a regression task. In machine learning, regression is used when the goal is to predict a continuous numeric value. In this case, overtime hours represent a continuous quantity that can vary based on different factors, such as the number of orders. Regression models establish a relationship between independent variables (in this case, the number of orders) and a dependent variable (the predicted overtime hours). This allows the model to provide estimates of overtime hours for different order quantities, making it suitable for scenarios where the output is a continuous range rather than discrete categories, which is characteristic of classification tasks. Classification, on the other hand, involves assigning discrete labels to inputs (e.g., determining if an email is spam or not), while clustering is used for grouping similar data points without predefined labels. Anomaly detection is focused on identifying rare items or patterns that differ significantly from the majority of the data. Thus, these options are not applicable to the scenario described.