Elasticsearch did-you-mean example

This repository contains everything you need to try out the did-you-mean functionality provided by the Elasticsearch phrase suggester.

Getting started

To run the example, you'll need to have npm and docker.

First install dependencies:

npm install

Then up the docker environment and import the mapping and some sample data:

npm start

This will open up the kibana dashboard, where you can then try out some queries.

Running suggest queries

After setting up the environment, you can run queries directly in Kibana. You can access the dev tools by going to this url.

For example, you can run the following query to get some suggestions:

GET library/book/_search
{
  "suggest": {
    "text": "prisoner of axaban",
    "phrase_suggester": {
      "phrase": {
        "field": "title.shingle",
        "confidence": 0.0,
        "direct_generator": [
          {
            "field": "title.shingle"
          }
        ]
      }
    }
  }
}

To try this out with a collate query, run the following query:

GET library/book/_search
{
  "suggest": {
    "text": "prisoner of axaban",
    "simple_phrase": {
      "phrase": {
        "field": "title.shingle",
        "confidence": 0.0,
        "direct_generator": [
          {
            "field": "title.shingle"
          }
        ],
        "collate": {
          "query": {
            "source": {
              "match": {
                "title": {
                  "query": "{{suggestion}}",
                  "fuzziness": "1",
                  "operator": "and"
                }
              }
            }
          },
          "prune": "true"
        }
      }
    }
  }
}