In Elasticsearch, an analyzer is applied to text during the indexing and search process. When you index a document, the text is first passed through the analyzer to generate a list of tokens that represent the content of the document. These tokens are then stored in the inverted index, which is used to quickly retrieve documents that match a particular search query.
When you perform a search, the search query is also passed through the same analyzer to generate a list of tokens. These tokens are then compared against the tokens in the inverted index to determine which documents match the query.
By applying the same analyzer to both the indexed documents and the search query, Elasticsearch ensures that the same tokenization and normalization rules are applied consistently. This helps to improve the accuracy and relevance of search results.
Note that you can specify a different analyzer for each field in your Elasticsearch index. This allows you to apply different tokenization and normalization rules to different types of content, depending on your specific requirements.