Elastic Learned Sparse EncodeR - or ELSER - is an NLP model trained by Elastic that enables you to perform semantic search by using sparse vector representation. Instead of literal matching on search terms, semantic search retrieves results based on the intent and the contextual meaning of a search query.

The instructions in this tutorial shows you how to use ELSER to perform semantic search on your data.

Requirements

To perform semantic search by using ELSER, you must have the NLP model deployed in your cluster. Refer to the ELSER documentation to learn how to download and deploy the model.

Create the index mapping

First, the mapping of the destination index - the index that contains the tokens that the model created based on your text - must be created. The destination index must have a field with the sparse_vector or rank_features field type to index the ELSER output.

curl -X PUT "${ES_URL}/my-index" \
-H "Authorization: ApiKey ${API_KEY}" \
-H "Content-Type: application/json" \
-d'
{
  "mappings": {
    "properties": {
      "content_embedding": { 
        "type": "sparse_vector" 
      },
      "content": { 
        "type": "text" 
      }
    }
  }
}
'

To learn how to optimize space, refer to the Saving disk space by excluding the ELSER tokens from document source section.

Create an ingest pipeline with an inference processor

Create an ingest pipeline with an inference processor to use ELSER to infer against the data that is being ingested in the pipeline.

curl -X PUT "${ES_URL}/_ingest/pipeline/elser-v2-test" \
-H "Authorization: ApiKey ${API_KEY}" \
-H "Content-Type: application/json" \
-d'
{
  "processors": [
    {
      "inference": {
        "model_id": ".elser_model_2",
        "input_output": [ <1>
          {
            "input_field": "content",
            "output_field": "content_embedding"
          }
        ]
      }
    }
  ]
}
'

Load data

In this step, you load the data that you later use in the inference ingest pipeline to extract tokens from it.

Use the msmarco-passagetest2019-top1000 data set, which is a subset of the MS MARCO Passage Ranking data set. It consists of 200 queries, each accompanied by a list of relevant text passages. All unique passages, along with their IDs, have been extracted from that data set and compiled into a tsv file.

Download the file and upload it to your cluster using the Data Visualizer in the Machine Learning UI. Assign the name id to the first column and content to the second column. The index name is test-data. Once the upload is complete, you can see an index named test-data with 182469 documents.

Ingest the data through the inference ingest pipeline

Create the tokens from the text by reindexing the data throught the inference pipeline that uses ELSER as the inference model.

curl -X POST "${ES_URL}/_reindex?wait_for_completion=false" \
-H "Authorization: ApiKey ${API_KEY}" \
-H "Content-Type: application/json" \
-d'
{
  "source": {
    "index": "test-data",
    "size": 50 
  },
  "dest": {
    "index": "my-index",
    "pipeline": "elser-v2-test"
  }
}
'

The call returns a task ID to monitor the progress:

curl -X GET "${ES_URL}/_tasks/<task_id>" \
-H "Authorization: ApiKey ${API_KEY}" \

You can also open the Trained Models UI, select the Pipelines tab under ELSER to follow the progress.

Semantic search by using the text_expansion query

To perform semantic search, use the text_expansion query, and provide the query text and the ELSER model ID. The example below uses the query text "How to avoid muscle soreness after running?", the content_embedding field contains the generated ELSER output:

curl -X GET "${ES_URL}/my-index/_search" \
-H "Authorization: ApiKey ${API_KEY}" \
-H "Content-Type: application/json" \
-d'
{
   "query":{
      "text_expansion":{
         "content_embedding":{
            "model_id":".elser_model_2",
            "model_text":"How to avoid muscle soreness after running?"
         }
      }
   }
}
'

The result is the top 10 documents that are closest in meaning to your query text from the my-index index sorted by their relevancy. The result also contains the extracted tokens for each of the relevant search results with their weights.

"hits": {
  "total": {
    "value": 10000,
    "relation": "gte"
  },
  "max_score": 26.199875,
  "hits": [
    {
      "_index": "my-index",
      "_id": "FPr9HYsBag9jXmT8lEpI",
      "_score": 26.199875,
      "_source": {
        "content_embedding": {
          "muscular": 0.2821541,
          "bleeding": 0.37929374,
          "foods": 1.1718726,
          "delayed": 1.2112266,
          "cure": 0.6848574,
          "during": 0.5886185,
          "fighting": 0.35022718,
          "rid": 0.2752442,
          "soon": 0.2967024,
          "leg": 0.37649947,
          "preparation": 0.32974035,
          "advance": 0.09652356,
          (...)
        },
        "id": 1713868,
        "model_id": ".elser_model_2",
        "content": "For example, if you go for a run, you will mostly use the muscles in your lower body. Give yourself 2 days to rest those muscles so they have a chance to heal before you exercise them again. Not giving your muscles enough time to rest can cause muscle damage, rather than muscle development."
      }
    },
    (...)
  ]
}

To learn about optimizing your text_expansion query, refer to Optimizing the search performance of the text_expansion query.

Combining semantic search with other queries

You can combine text_expansion with other queries in a compound query. For example using a filter clause in a Boolean query or a full text query which may or may not use the same query text as the text_expansion query. This enables you to combine the search results from both queries.

The search hits from the text_expansion query tend to score higher than other Elasticsearch queries. Those scores can be regularized by increasing or decreasing the relevance scores of each query by using the boost parameter. Recall on the text_expansion query can be high where there is a long tail of less relevant results. Use the min_score parameter to prune those less relevant documents.

curl -X GET "${ES_URL}/my-index/_search" \
-H "Authorization: ApiKey ${API_KEY}" \
-H "Content-Type: application/json" \
-d'
{
  "query": {
    "bool": { 
      "should": [
        {
          "text_expansion": {
            "content_embedding": {
              "model_text": "How to avoid muscle soreness after running?",
              "model_id": ".elser_model_2",
              "boost": 1 
            }
          }
        },
        {
          "query_string": {
            "query": "toxins",
            "boost": 4 
          }
        }
      ]
    }
  },
  "min_score": 10 
}
'

Optimizing performance

Saving disk space by excluding the ELSER tokens from document source

The tokens generated by ELSER must be indexed for use in the text_expansion query. However, it is not necessary to retain those terms in the document source. You can save disk space by using the source exclude mapping to remove the ELSER terms from the document source.

The mapping that excludes content_embedding from the _source field can be created by the following API call:

curl -X PUT "${ES_URL}/my-index" \
-H "Authorization: ApiKey ${API_KEY}" \
-H "Content-Type: application/json" \
-d'
{
  "mappings": {
    "_source": {
      "excludes": [
        "content_embedding"
      ]
    },
    "properties": {
      "content_embedding": {
        "type": "sparse_vector"
      },
      "content": {
        "type": "text"
      }
    }
  }
}
'

Further reading

• How to download and deploy ELSER
• ELSER limitation
• Improving information retrieval in the Elastic Stack: Introducing Elastic Learned Sparse Encoder, our new retrieval model

Interactive example

• The elasticsearch-labs repo has an interactive example of running ELSER-powered semantic search using the Elasticsearch Python client.