# Use distance on dense vectors in relevance score (at query time)
**URL:** https://discuss.elastic.co/t/use-distance-on-dense-vectors-in-relevance-score-at-query-time/217012
**Category:** Elasticsearch
**Created:** [January 29, 2020, 1:33pm UTC](https://discuss.elastic.co/t/use-distance-on-dense-vectors-in-relevance-score-at-query-time/217012 "2020-01-29T13:33:53Z")
**Posts on this page:** 4
**Page:** 1
### Author:  [@thomas-chauvet](https://discuss.elastic.co/u/thomas-chauvet)
#### Post date: [January 29, 2020, 1:33pm UTC](https://discuss.elastic.co/t/use-distance-on-dense-vectors-in-relevance-score-at-query-time/217012/1 "2020-01-29T13:33:53Z")
I use elasticsearch to combine different things:
- search in text
- score based on dense vector (cosine similarity)
I use a query with `function_score`. The first part is the search in the text (giving a score) and THEN a script is applied to compute cosine similarity.
My problem is that the cosine similarity is not computed during the query phase and my search in the text act as a pre-filter. I will always obtain results linked with the text search even if the cosine similarity is better.
This is the standard behavior of `function_score` according the [doc](https://www.elastic.co/guide/en/elasticsearch/reference/current/query-dsl-function-score-query.html):
> The function\_score allows you to modify the score of documents that are retrieved by a query. This can be useful if, for example, a score function is computationally expensive and it is sufficient to compute the score on a filtered set of documents.
I would like to compute the cosine similarity at query time and this score will be combined with the text search (with as much importance).
Thanks !
You will find a gist [here](https://gist.github.com/thomas-chauvet/0a8317296800ecd3a92cb926e260fe7d) describing the problem with a "real" example.
_Note: this post is also on [stackoverflow](https://stackoverflow.com/questions/59967308/elasticsearch-use-distance-on-dense-vectors-in-relevance-score-at-query-time)_
---
### Author:  [@mayya](https://discuss.elastic.co/u/mayya)
#### Post date: [January 31, 2020, 8:37pm UTC](https://discuss.elastic.co/t/use-distance-on-dense-vectors-in-relevance-score-at-query-time/217012/2 "2020-01-31T20:37:26Z")
Hi there!
Did I understand your intention correctly: you want to go through **all** documents, and apply cosine similarity function to them. Then you also have a query and for the documents that match a query, you want to calculate score for this query. Then you want to combine these two scores: from a cosine similarity and a query?
Currently, you can do that with a bool query using `should` clauses like this:
```auto
GET my_index/_search
{
"query": {
"bool": {
"should" : [
{
"match": {
"my_text": {
"query": "abc"
}
}
},
{
"script_score" : {
"query" : {"match_all" : {}},
"script" : {
"source": "50 * cosineSimilarity(params.query_vector, doc['my_vector']) + 1.0",
"params": {
"query_vector": [0, 0, 1]
}
}
}
}
]
}
}
}
```
This will give you a sum of scores: score1 + score2. You can also apply `boost` for any query.
We also have a plan to develop a [compound query](https://github.com/elastic/elasticsearch/issues/42811#issuecomment-524854701) that will give you an option to combine scores of queries not only through `sum` option. But this is not available yet.
---
### Author:  [@thomas-chauvet](https://discuss.elastic.co/u/thomas-chauvet)
#### Post date: [February 4, 2020, 3:21pm UTC](https://discuss.elastic.co/t/use-distance-on-dense-vectors-in-relevance-score-at-query-time/217012/3 "2020-02-04T15:21:21Z")
Thanks you, it was exactly what I was looking for. I didn't now that I could combine queries like this.
If you want to copy your anser on [stackoverlow](https://stackoverflow.com/questions/59967308/elasticsearch-use-distance-on-dense-vectors-in-relevance-score-at-query-time) 🙂
---
### Author:  [@system](https://discuss.elastic.co/u/system)
#### Post date: [March 3, 2020, 3:21pm UTC](https://discuss.elastic.co/t/use-distance-on-dense-vectors-in-relevance-score-at-query-time/217012/4 "2020-03-03T15:21:29Z")
This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.