IMPORTANT: No additional bug fixes or documentation updates will be released for this version. For the latest information, see the current release documentation.
« Rule query Term-level queries »
Elastic Docs ›Elasticsearch Guide [8.13] ›Query DSL ›Specialized queries

Weighted tokens query

edit

Weighted tokens query

edit

This functionality is in technical preview and may be changed or removed in a future release. Elastic will work to fix any issues, but features in technical preview are not subject to the support SLA of official GA features.

The weighted tokens query requires a list of token-weight pairs that are sent in with a query rather than calculated using a natural language processing model. These token pairs are then used in a query against a sparse vector or rank features field.

Weighted tokens queries are useful when you want to use an external query expansion model, or quickly prototype changes without reindexing a new model.

Example request

edit
POST _search
{
  "query": {
    "weighted_tokens": {
      "query_expansion_field": {
        "tokens": {"2161": 0.4679, "2621": 0.307, "2782": 0.1299, "2851": 0.1056, "3088": 0.3041, "3376": 0.1038, "3467": 0.4873, "3684": 0.8958, "4380": 0.334, "4542": 0.4636, "4633": 2.2805, "4785": 1.2628, "4860": 1.0655, "5133": 1.0709, "7139": 1.0016, "7224": 0.2486, "7387": 0.0985, "7394": 0.0542, "8915": 0.369, "9156": 2.8947, "10505": 0.2771, "11464": 0.3996, "13525": 0.0088, "14178": 0.8161, "16893": 0.1376, "17851": 1.5348, "19939": 0.6012},
        "pruning_config": {
          "tokens_freq_ratio_threshold": 5,
          "tokens_weight_threshold": 0.4,
          "only_score_pruned_tokens": false
        }
      }
    }
  }
}

Top level parameters for weighted_token

edit
<tokens>

(Required, dictionary) A dictionary of token-weight pairs.

pruning_config

(Optional, object) Optional pruning configuration. If enabled, this will omit non-significant tokens from the query in order to improve query performance. Default: Disabled.

Parameters for <pruning_config> are:

tokens_freq_ratio_threshold
(Optional, integer) Tokens whose frequency is more than tokens_freq_ratio_threshold times the average frequency of all tokens in the specified field are considered outliers and pruned. This value must between 1 and 100. Default: 5.
tokens_weight_threshold
(Optional, float) Tokens whose weight is less than tokens_weight_threshold are considered nonsignificant and pruned. This value must be between 0 and 1. Default: 0.4.
only_score_pruned_tokens
(Optional, boolean) If true we only input pruned tokens into scoring, and discard non-pruned tokens. It is strongly recommended to set this to false for the main query, but this can be set to true for a rescore query to get more relevant results. Default: false.

The default values for tokens_freq_ratio_threshold and tokens_weight_threshold were chosen based on tests using ELSER that provided the most optimal results.

Example weighted tokens query with pruning configuration and rescore

edit

The following example adds a pruning configuration to the text_expansion query. The pruning configuration identifies non-significant tokens to prune from the query in order to improve query performance.

Token pruning happens at the shard level. While this should result in the same tokens being labeled as insignificant across shards, this is not guaranteed based on the composition of each shard. Therefore, if you are running text_expansion with a pruning_config on a multi-shard index, we strongly recommend adding a Rescore filtered search results function with the tokens that were originally pruned from the query. This will help mitigate any shard-level inconsistency with pruned tokens and provide better relevance overall.

GET my-index/_search
{
   "query":{
      "weighted_tokens": {
      "query_expansion_field": {
        "tokens": {"2161": 0.4679, "2621": 0.307, "2782": 0.1299, "2851": 0.1056, "3088": 0.3041, "3376": 0.1038, "3467": 0.4873, "3684": 0.8958, "4380": 0.334, "4542": 0.4636, "4633": 2.2805, "4785": 1.2628, "4860": 1.0655, "5133": 1.0709, "7139": 1.0016, "7224": 0.2486, "7387": 0.0985, "7394": 0.0542, "8915": 0.369, "9156": 2.8947, "10505": 0.2771, "11464": 0.3996, "13525": 0.0088, "14178": 0.8161, "16893": 0.1376, "17851": 1.5348, "19939": 0.6012},
        "pruning_config": {
          "tokens_freq_ratio_threshold": 5,
          "tokens_weight_threshold": 0.4,
          "only_score_pruned_tokens": false
        }
      }
    }
   },
   "rescore": {
      "window_size": 100,
      "query": {
         "rescore_query": {
            "weighted_tokens": {
              "query_expansion_field": {
                "tokens": {"2161": 0.4679, "2621": 0.307, "2782": 0.1299, "2851": 0.1056, "3088": 0.3041, "3376": 0.1038, "3467": 0.4873, "3684": 0.8958, "4380": 0.334, "4542": 0.4636, "4633": 2.2805, "4785": 1.2628, "4860": 1.0655, "5133": 1.0709, "7139": 1.0016, "7224": 0.2486, "7387": 0.0985, "7394": 0.0542, "8915": 0.369, "9156": 2.8947, "10505": 0.2771, "11464": 0.3996, "13525": 0.0088, "14178": 0.8161, "16893": 0.1376, "17851": 1.5348, "19939": 0.6012},
                "pruning_config": {
                  "tokens_freq_ratio_threshold": 5,
                  "tokens_weight_threshold": 0.4,
                  "only_score_pruned_tokens": true
                }
              }
            }
         }
      }
   }
}
« Rule query Term-level queries »