Training Q

Question & Answer
, ,
1

Hello, is there more resources on how to train Q to give similar outputs to the ones that received positive feedback?

I am confused about training Q properly. I am trying to make it simple for my non technical users to get insights about the data they want with natural language. So far I have found ways to structure prompts in a complicated way to answer. Are there more resources about how to train Q?

2

Hello dduranm,

First of all Welcome to the QuickSight Community.

I think there is no direct way to train but rather follow the best practices and provide feedback for questions and if necessary define the unrecognised term.

Please refer to some of the links below.

  1. Supported types of questions by Q - Asking questions with Amazon QuickSight Q - Amazon QuickSight
  2. Providing Feedback - Providing feedback about Amazon QuickSight Q topics - Amazon QuickSight
  3. Correcting Wrong answers - defining unrecognized terms - Correcting wrong answers provided by Amazon QuickSight Q - Amazon QuickSight
  4. Following best practices to enable end user to use NL querying -Best practices for enabling business users to answer questions about data using natural language in Amazon QuickSight | AWS Business Intelligence Blog

Hope this helps.
Cheers,
Deep

3

Hi @dduranm,
It’s been awhile since we last heard from you. Were you able to check out the information shared above and was it useful for your case?

If we don’t hear back from you within 3 business days, I’ll go ahead and mark the solution.

Thank you!