When enriching chatlogs we want to look at each sentence level. The proposed solution here is to split each sentence to seperate documents. Want this to be able to support as an array field inside a JSON file.
"text": ["hello", "how are you today?]
How i need to do it today:
"text2": "how are you today?"
Challenge here is that i have to select each fields to be enriched.
Why is it useful?
|Who would benefit from this IDEA?||as a user of the discovery service i want to upload only 1 json file containing the necessary fields that needs to be enriched instead of seperate files.|
How should it work?