# How to normalize similarity score using 'more\_like\_this'

**URL:** <https://discuss.elastic.co/t/how-to-normalize-similarity-score-using-more-like-this/249721>\
**Category:** Elasticsearch\
**Created:** [September 24, 2020, 12:54am UTC](https://discuss.elastic.co/t/how-to-normalize-similarity-score-using-more-like-this/249721 "2020-09-24T00:54:11Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![Sangmin\_Lee](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/sangmin_lee/32/45032_2.png) [@Sangmin\_Lee](https://discuss.elastic.co/u/Sangmin_Lee)\
**Post date:** [September 24, 2020, 12:54am UTC](https://discuss.elastic.co/t/how-to-normalize-similarity-score-using-more-like-this/249721/1 "2020-09-24T00:54:11Z")

</div>

Hi.  
I'm using 'more like this' to get the similarity of the given text.  
As you know, the score is not normalized.  
I want to get a normalized value [0, 1] so that I can cut out documents below the threshold score.

I thought if I can get the score of the query text itself,  
I could normalize it using 'matched document score/query text score'  
Is there any way to get score of query text itself?  
Or is there another way to normalize the similarity score?

Thanks for reading!

---

<div class="post-metadata">

**Author:** ![system](https://us1.discourse-cdn.com/elastic/original/3X/1/a/1ac57faf039f6b580b3f104ef42a2a89e41014de.png) [@system](https://discuss.elastic.co/u/system)\
**Post date:** [October 22, 2020, 12:54am UTC](https://discuss.elastic.co/t/how-to-normalize-similarity-score-using-more-like-this/249721/2 "2020-10-22T00:54:16Z")

</div>

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