In supervised machine learning, the key difference between classification and regression lies in the nature of the output they predict. Classification algorithms are used to assign data points to one of several predefined categories or classes, making it suitable for tasks like spam detection, where an email is classified as either "spam" or "not spam." On the other hand, regression algorithms predict continuous values, such as forecasting the price of a house based on features like size, location, and number of rooms. While classification answers "which category?" regression answers "how much?" or "what value?".