In this course, employee attrition prediction means using employee-related data to build a model that estimates who is more likely to leave. The emphasis is on learning the practical workflow around that task, including preparing data, exploring patterns, and testing the model.

You would use attrition prediction when you want a structured way to assess quitting risk from multiple employee factors instead of relying on one signal by itself. In the course, it is treated as a repeatable machine learning task that turns workforce data into a prediction problem.

It fits into the build-and-test phase of a broader HR data workflow, after data is collected and before predictions are applied in practice. The course shows that good prediction depends on cleaning data, shaping features, and evaluating model behavior as part of one connected process.