multi_match

A multi_match query performs a lexical search over multiple searchable text fields. The search query is analyzed before performing search.

If you would like to search over a semantic-indexed field, consider a semantic and knn search operators instead.

An operator similar to match but able to search multiple fields at once:

{
  "query": {
    "multi_match": {
      "fields": ["<field-name>", "<field-name>"],
      "query": "<search-query>",
      "type": "best_fields" //(1) 
    }
  }
}

1. optional

Where:

• <field-name>: is an existing field marked as searchable.
• <search-query>: a search query string.
• type: optional, possible values: "best_fields", "most_fields", default "best_fields". The way field scores are combined when multiple of them are matched at once.

Wildcard fields

multi_match query also supports wildcard fields, so you can make requests like this example:

{
  "query": {
    "multi_match": {
      "fields": ["desc_*"],
      "query": "cookies" 
    }
  }
}

This query will perform a multi_match over all fields starting with desc_ prefix.

Types of multi_match queries

The multi_match query has multiple behaviors depending on the type parameter:

1. best_fields (default): Finds all documents which match any field, but only the _score from the best matching field is used.
2. most_fields: Finds all documents which match any field, and combines the _score from each field.

best_fields

An example for best_fields query is searching over title and description fields, where a match over a title field might be more important than match over the desription field.

The full schema of the query:

{
  "query": {
    "multi_match": {
      "fields": ["title", "description"],
      "query": "cookies",
      "type": "best_fields",
      "tie_breaker": 0.3
    }
  }
}

The best_fields query is an alias for a dis_max query, which generates multiple per-field match queries and wraps them the following way:

{
  "query": {
    "dis_max": {
      "queries": [
        {"match": {"title": "cookies"}},
        {"match": {"description": "cookies"}}
      ],
      "tie_breaker": 0.3
    }
  }
}

Normally the best_fields type uses the score of the single best matching field, but if tie_breaker is specified, then it calculates the score as follows:

1. the score from the best matching field
2. plus tie_breaker * _score for all other matching fields

most_fields

The most_fields query type is useful when searching over similar fields analyzed in a different manner, for example in different languages:

schema:
  my-index:
    fields:
      title_english:
        type: text
        search:
          lexical:
            analyze: english
      title_spanish:
        type: text
        search:
          lexical:
            analyze: spanish

So the following query can be used:

{
  "query": {
    "multi_match": {
      "fields": ["title_english", "title_spanish"],
      "query": "cookies",
      "type": "most_fields"
    }
  }
}

Internally the most_fields query is implemented as a bool query over multiple per-field match queries:

{
  "query": {
    "bool": {
      "should": [
        {"match": {"title_english": "cookies"}},
        {"match": {"title_spanish": "cookies"}}
      ]
    }
  }
}