Suggestions for a better text based query - using match_phrase but it's too precise

Elastic Stack Elasticsearch
1

Hi there,

My title probably doesn't elaborate that well.

We have a significant dataset, where our end-users can search base don a job_title.

Currently I'm using:

bool
should
match_phrase

We also include "minimum_should_match" : 1" in the query.

But on occassions we get results far too narrow. For example, we in theory have 1000's of matches for this .

Yet, we only return two e.g.:

'Woodshop Manager And Furniture Designer'
and
'Furniture Manager, Designer, Buyer'

So i'm hoping for some expert advice from the forum, what should I look at for the cause of this issue? It's not matching 'exact' which could be a cause of the issue, as the second result doesn't have the exact match 'Furniture Designer' - so is there a better query instead of match_phrase I can use?

I can't use query_string or simple_string as we are using a join in our schema to allow for multiple queries to be dynamically added.

Any guidance appreciated.

2

Gentle bump!

3

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