Labels

Overview

Labels are the correct answers associated with training data.

In supervised machine learning, labels tell the model what outcome it should learn to predict.

A dataset used to identify spam emails, for example, may contain labels such as “spam” and “not spam.”

By learning from labeled examples, the model gradually improves its ability to make predictions.

Why It Matters

Labels provide the feedback needed for learning.

Without labels, supervised machine learning models would not know whether their predictions are correct.

Real-World Example

An email provider trains a spam filter using messages labeled as spam or not spam.

The model learns patterns associated with each label and uses them to classify future emails.

Related Concepts

• Dataset
• Features
• Classification
• Regression
• Supervised Learning