Microsoft Azure AI Fundamentals (AI-900) Practice Exam

What type of model would you use to predict a student's likelihood of completing a university course?

1. Clustering

2. Regression

3. Classification

4. Reinforcement Learning

The correct answer is: Classification

Using a classification model to predict a student's likelihood of completing a university course is appropriate because classification is specifically designed for situations where the outcome or target variable is categorical. In this case, the outcome can be framed as either "will complete" or "will not complete" the course, thus creating two distinct classes. Classification models analyze data features—such as attendance, grades, prior academic performance, and engagement—to estimate the probability that a student falls into one of those categories. These models can help educators and administrators identify at-risk students and implement proactive measures to support their success. Other approaches such as clustering are used for grouping data points based on similarity without predefined categories, which does not suit the need for a clear binary outcome. Regression, on the other hand, is typically used to predict continuous numerical values rather than categorical outcomes. Lastly, reinforcement learning is more applicable for scenarios where an agent interacts with an environment and learns to make decisions based on rewards and penalties, rather than predicting outcomes based on existing data patterns in a fixed dataset.