Supervised learning
Supervised learning is a machine learning technique in which a system learns to solve problems from a large collection of examples that already have the correct answer.
On this page
Source content checked . English translation published 2026-10-27.
In business automation
Classifying an inquiry or predicting churn requires examples that have already been labeled with the correct answer. Supervised learning is a technique in which a system learns from such a collection. In the guide, the term appears alongside machine learning and reinforcement learning, without claiming that any particular tool trains such a model.
Examples
These examples illustrate a possible workflow. Adapt the connections and rules to your own systems.
Labeling past inquiries before evaluating classification
A team labels a thousand past inquiries by department. They set aside some examples for evaluation, and only then check whether a model that classifies a new inquiry succeeds on examples that were not included in training.
Relevant tools
- Replicate
- A service for running open-source models. Supervised learning is a training method. This does not claim that the tool trains a model for you.
What to check
- Training and evaluating on the same labeled examples produces a high score that does not hold up on new inquiries.