A text analyzer transforms language-specific text into tokens suitable for indexing and searching. Language analyzers can remove stop words and reduce related word forms through stemming. section 7.3.7 section 7.0
Language analyzers support practical search behavior through configurable processing:
- Select analyzers for languages including English, German, Hindi, Spanish, French, and many others. section 7.3.7
- Configure custom stop-word lists to avoid indexing common or unwanted terms. section 7.3.7
- Use stemming to map forms such as author, authors, authoring, and authored to author. section 7.3.7
- Exclude words such as authority and authorization when stemming would incorrectly reduce them to author. section 7.3.7
Testing a custom English analyzer
Create an analyzer with stop words ["a", "an", "is", "and", "for"], then send "A dog is for a life" to the _analyze API. The resulting tokens are dog and life, showing how configuration changes indexed terms.
