4

In the QnA Maker Service provided by the Microsoft's Cognitive Services, we can train our KB (Knowledge Base) by manually inserting the QnA pairs.

Is there a way to automate this process so that we can always keep our KB up-to-date please? There is an API documentation but i cannot find one for this purpose.

Please advise. Thanks all in advance.

wmtechnologies
  • 83
  • 5
  • See https://qnamaker.ai/Documentation/ApiReference. There are APIs to programmatically manage your KB. – Prashant Choudhari Jul 04 '17 at 04:46
  • if i am not wrong, question is related to training model, when link has been established between question and answer while testing within QnA portal. I am not sure, but are you looking for something like this [Train API_Active learning](https://learn.microsoft.com/en-us/azure/cognitive-services/qnamaker/how-to/improve-knowledge-base#train-api) – wannabdev Jun 05 '20 at 18:21

1 Answers1

1

V2 of the QnA Maker API provides the following to help you programmatically train the knowledge base:

  • POST Create Knowledge Base
  • DELETE Delete Knowledge Base
  • GET Download Alterations
  • GET Download Knowledge Base
  • POST Generate answer
  • PUT Publish Knowledge Base
  • PATCH Train Knowledge Base
  • PATCH Update Alterations
  • PATCH Update Knowledge Base

For more information, see the documentation here: https://westus.dev.cognitive.microsoft.com/docs/services/58994a073d9e04097c7ba6fe/operations/58994a073d9e041ad42d9baa

Luke Bayler
  • 11
  • 5