# Supervised learning · Glossary
A machine-learning approach in which a model learns from examples that are labelled with the correct answer, such as past claims marked as fraudulent or not.
[Glossary](/glossary) · Machine learning and MLOps

# Supervised learning

     A machine-learning approach in which a model learns from examples that are labelled with the correct answer, such as past claims marked as fraudulent or not. The trained model then predicts that label for new cases.

## Why it matters

     Most production machine learning in business is supervised: classification, scoring and forecasting problems where historical outcomes are known. Its performance depends directly on the quality, coverage and relevance of those labelled examples.

     Historical labels also carry historical decisions. If past approvals or investigations were inconsistent or biased, a supervised model will learn those patterns unless the data and evaluation are designed to detect them.

## In practice

     For example, a UK water utility might train a supervised model on several years of inspection records, labelled with whether a pipe section later failed, to prioritise which sections receive inspections first.

## Where Rodan fits

     Rodan builds and validates supervised models for operational decisions in [AI and Decision Systems](https://rodan.io/what-we-build/ai-decision-systems) work. See also [the difference between AI, machine learning and data science](https://rodan.io/insights/the-difference-between-ai-machine-learning-and-data-science).

## Related terms

- [Machine learning (ML)](/glossary#machine-learning)

- [Unsupervised learning](/glossary#unsupervised-learning)

- [Training data](/glossary/training-data)

- [Ground truth](/glossary/ground-truth)

- [Overfitting](/glossary/overfitting)

- [Data labelling](/glossary/data-labelling)
HTML: https://rodan.io/glossary/supervised-learning