elasticsearchHow do I add synonyms to Elasticsearch?

Adding synonyms to Elasticsearch is a great way to improve the accuracy of your search results. Synonyms are words or phrases that have the same meaning as another word or phrase.

To add synonyms to Elasticsearch, you need to use the synonym_graph token filter. To do this, you must create an index with a settings and mapping file.

For example, the following settings and mapping file will create an index with the synonym_graph token filter:

PUT /my_index
{
  "settings": {
    "analysis": {
      "analyzer": {
        "my_synonyms": {
          "tokenizer": "standard",
          "filter": [
            "lowercase",
            "my_synonym_filter"
          ]
        }
      },
      "filter": {
        "my_synonym_filter": {
          "type": "synonym_graph",
          "synonyms": [
            "quick,fast",
            "slow,sluggish"
          ]
        }
      }
    }
  },
  "mappings": {
    "properties": {
      "text": {
        "type": "text",
        "analyzer": "my_synonyms"
      }
    }
  }
}

The synonyms array is where you can specify the synonyms you want to add. In this example, we have specified two sets of synonyms: quick and fast, and slow and sluggish.

Once you have created the index, you can use the synonym_graph token filter to analyze text and add the synonyms you have specified. For example:

POST /my_index/_analyze
{
  "analyzer": "my_synonyms",
  "text": "I'm looking for something quick"
}

This will return the following output:

{
  "tokens": [
    {
      "token": "i'm",
      "start_offset": 0,
      "end_offset": 3,
      "type": "<ALPHANUM>",
      "position": 0
    },
    {
      "token": "looking",
      "start_offset": 4,
      "end_offset": 11,
      "type": "<ALPHANUM>",
      "position": 1
    },
    {
      "token": "for",
      "start_offset": 12,
      "end_offset": 15,
      "type": "<ALPHANUM>",
      "position": 2
    },
    {
      "token": "something",
      "start_offset": 16,
      "end_offset": 25,
      "type": "<ALPHANUM>",
      "position": 3
    },
    {
      "token": "quick",
      "start_offset": 26,
      "end_offset": 31,
      "type": "<ALPHANUM>",
      "position": 4
    },
    {
      "token": "fast",
      "start_offset": 26,
      "end_offset": 31,
      "type": "SYNONYM",
      "position": 4
    }
  ]
}

As you can see, the synonym_graph token filter has added the synonym fast to the text.

Code explanation

**

1. PUT /my_index - Creates the index with settings and mapping file
2. "tokenizer": "standard" - Specifies the tokenizer to use
3. "filter": [ "lowercase", "my_synonym_filter" ] - Specifies the filter to use
4. "type": "synonym_graph" - Specifies the type of token filter to use
5. "synonyms": [ "quick,fast", "slow,sluggish" ] - Specifies the synonyms to add
6. POST /my_index/_analyze - Analyzes the text and adds the synonyms

## Helpful links

• Elasticsearch Synonym Token Filter Documentation

Edit this code on GitHub