Microsoft Azure AI Fundamentals (AI-900) Practice Exam

Which model type is best suited for predicting continuous outcomes, like sales numbers?

1. Classification

2. Regression

3. Clustering

4. Dimensionality reduction

The correct answer is: Regression

When predicting continuous outcomes such as sales numbers, regression is the most suitable model type. Regression analysis specifically focuses on modeling the relationship between a dependent variable (in this case, sales numbers) and one or more independent variables. This allows for the prediction of continuous values based on input data. Unlike classification, which is used to categorize data into discrete classes or labels, regression is designed to provide a quantitative output, making it ideal for situations where the goal is to forecast or estimate continuous metrics. Clustering, on the other hand, involves grouping data points based on similarity, which does not directly apply to predicting specific numerical outcomes. Dimensionality reduction techniques aim to simplify datasets while preserving important information, but they do not inherently provide predictions for continuous values either. Therefore, regression stands out as the appropriate choice for scenarios that require predicting continuous outcomes like sales figures.