I have been trying to model the small talk model of api.ai in LUIS, but it is getting out of hand quickly. Any suggestions on how to do this would be appreciated.
are you busy, are you ok, are you there can you advise, can you hear me, can you help, can you kill, can you move, can you speak, can you talk, can you see me, can you undress, can you wave answer my question, ask me a question be clever, be my friend, be quiet, be stupid change your do it find boyfriend girlfriend.
These are few examples of small talk conversations.
api.ai has a domain called smalltalk which responds to such conversations. My question is how do I model these in LUIS.
I've created an easy and customizable way to handle smalltalk using the Microsoft Bot Framework, QnA maker and LUIS. It is a port over of API.AI's smalltalk. You can find the source code here: https://github.com/alyssaong1/botframework-smalltalk
How it works: You will need to create a QnA maker service at https://qnamaker.ai and upload the knowledge base full of utterances to the QnA maker service. Then, upload all the utterances to an intent called "smalltalk" in LUIS. When the user's utterance falls under this intent, pass the utterance to the QnA maker to be handled. More details in the repository.
Understand that it's been a while since you've posted this question, but I'll just leave this here for those who are still facing the same issue!
If you love us? You can donate to us via Paypal or buy me a coffee so we can maintain and grow! Thank you!
Donate Us With