Compare two columns with fuzziness or slops

Elastic Stack Elasticsearch
1

Hi,
I'm trying to compare two columns such as address1 and address2, but also want to apply fuzziness or slop values. I could get the exact match in the following in Kibana.

GET /my_index/_search
{
"query": {
"bool": {
"must": [
{
"script": {
"script": "doc['address1.keyword'].value==doc['address2.keyword'].value"
}
}
]
}
}
}

But I couldn't apply fuzziness or slop values on script.
Is there any (better) way to perform this task?

Thanks

2

that is not really possible with access doc values, as you cannot execute another analysis and search step after receiving those values.

Can you explain, what exactly is your use-case? Maybe there is another way to work this out, but I am totally lacking context and I am curious to know more :slight_smile:

3

Let's say I have a table whose two fields are address1 and address2. Not only that I want to find out how many times these two fields are exactly the same, but also (more importantly) when they are different, I want to find if they can be considered the same using fuzziness or slop values.
Ex: "Adam Street" vs "Adams Street" using fuzziness=1, "Adam Street" vs "Street Adam" using slop=2.
Any advice would be appreciated.
Thanks

4

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