Match and term in bool's should

Elastic Stack Elasticsearch
1

I'd like to search items that either have firstField (mapped as text) similar to a query string or secondField (mapped as an integer) equal to, say, 1.

After reading the docs, I understand the OR I am trying to implement can be done with a should clause inside a bool like so:

{
  "query": {
    "bool": {
      "should": [
        {"match": {"firstField":  "queryString"}},
        {"term":  {"secondField": 1}}
      ]
    }
  }
}

The results however show that secondField contributes far less (approximately 4 times) to the result than firstField does since their scoring algorithms are different. firstField effectively eclipses secondField, even though I want them to weigh the same on the final score.

My question is twofold:

  • is it even recommended to have a match inside a should (or any other bool sub-clause) since the result is non-boolean?
  • how can I make two clauses of different natures (in this case match and term) weigh the same on the final score (have the same scoring span)?
2

This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.