您现在访问的是微软AZURE全球版技术文档网站，若需要访问由世纪互联运营的MICROSOFT AZURE中国区技术文档网站，请访问 https://docs.azure.cn.
Skillset operations (Azure Search Service REST API)
A skillset is a collection of AI-powered algorithms, referred to as cognitive skills, used for natural language processing and other transformations on content that you are indexing in Azure Search. A skillset can include built-in skills, custom skills, or a combination of both. Examples of built-in skills include entity recognition, key phrase extraction, chunking text into logical pages, among others.
To use the skillset, reference it in an indexer and then run the indexer to import data, invoking transformations and enrichment on in-transit data, mapping the results to output fields to an index. A skillset is high-level standalone resource that exists on a level equivalent to indexes, indexers, and data sources, but it is operational only within indexer processing. As a high-level resource, you can design a skillset once, and then reference it in multiple indexers.
Operations on skillsets
- Azure Search Service REST
- Service limits in Azure Search
- Azure Search .NET library
- Cognitive search overview
- Quickstart: Try cognitive search
- Tutorial: Learn the cognitive search REST APIs
- How to define a skillset
- How to map fields
- How to define a custom interface
- Example: creating a custom skill
- Predefined skills