Unsupervised learning: Difference between revisions
| Line 11: | Line 11: | ||
====Related terms==== | ====Related terms==== | ||
[https://en.wiktionary.org/wiki/unsupervised unsupervised] | * [https://en.wiktionary.org/wiki/unsupervised unsupervised] | ||
[https://en.wiktionary.org/wiki/learning learning] | * [https://en.wiktionary.org/wiki/learning learning] | ||
[https://en.wiktionary.org/wiki/machine_learning machine learning] | * [https://en.wiktionary.org/wiki/machine_learning machine learning] | ||
[https://en.wiktionary.org/wiki/artificial_intelligence artificial intelligence] | * [https://en.wiktionary.org/wiki/artificial_intelligence artificial intelligence] | ||
Latest revision as of 19:16, 9 May 2026
English
[edit | edit source]Noun
[edit | edit source]unsupervised learning
A machine-learning approach in which an AI system is trained on a dataset that has not been labeled by people, usually to identify patterns, structures, or groupings in data without a specific, known output in mind.
- For example, a model may analyze a large, unlabeled dataset of photos containing horses and zebras. Without being told what a zebra is, the AI may discover patterns on its own, such as striped or solid appearances, and cluster images with similar features together.