Tokenization in Elasticsearch

Keep in mind that Elasticsearch, by default, tokenizes or analyzes fields that are of type text. As a result, strings consisting of two or more words may become separate fields when connected to Composer (for example, city names like Las Vegas). To disable this process and ensure that a string field is not analyzed, specify its type as keyword:

City: {
type: "keyword"