In machine learning, what is the difference between regression and classification?

In machine learning, what is the difference between regression and classification?

• A. Regression predicts continuous outcomes, while classification predicts categorical outcomes.
• B. Regression predicts categorical outcomes, while classification predicts continuous values.
• C. Regression and classification both predict continuous values.
• D. Regression and classification are unrelated to predicting outcomes.

Answer: (A)

The key idea is distinguishing the type of target you’re predicting. Regression is used when the goal is to predict a continuous numeric value, such as a house price or a temperature. Classification, on the other hand, is about assigning inputs to discrete categories, like labeling an email as spam or not spam, or identifying an image as a dog, cat, or bird. This difference shapes the whole approach: the loss functions and evaluation metrics differ — regression uses errors like mean squared error, while classification uses accuracy or metrics that account for class balance, often with probabilistic outputs and cross-entropy loss. Therefore, predicting continuous outcomes for regression and predicting categorical outcomes for classification is the correct distinction. The other statements misstate the idea by claiming either that both predict continuous values, or that regression handles categories, or that these tasks aren’t about predicting outcomes.

The key idea is distinguishing the type of target you’re predicting. Regression is used when the goal is to predict a continuous numeric value, such as a house price or a temperature. Classification, on the other hand, is about assigning inputs to discrete categories, like labeling an email as spam or not spam, or identifying an image as a dog, cat, or bird. This difference shapes the whole approach: the loss functions and evaluation metrics differ — regression uses errors like mean squared error, while classification uses accuracy or metrics that account for class balance, often with probabilistic outputs and cross-entropy loss. Therefore, predicting continuous outcomes for regression and predicting categorical outcomes for classification is the correct distinction. The other statements misstate the idea by claiming either that both predict continuous values, or that regression handles categories, or that these tasks aren’t about predicting outcomes.