2

I try to get a scoring-value or a value for confidence of an intent using amazon-lex (PostText) but there is no response-element in json-file at all. (https://docs.aws.amazon.com/de_de/lex/latest/dg/API_runtime_PostText.html) Is there any level or threshold for a correct identification of an intent which is used by amazon-lex?

Thanks for your help.

Yannick Theisen
  • 21
  • 1
  • 1
    That would be a nice feature, but to my knowledge, there is no way for us to get that information. From what I've seen with Lex/Alexa, Amazon keeps the natural language understanding (NLU) part tightly concealed from devs, and we are forced to test a lot, and trust in the learning that occurs to improve the intents and slots identification. – Jay A. Little Feb 15 '19 at 11:31

1 Answers1

2

As of now, this feature is not available in Amazon Lex, we only get the matched intent as the response for post_text api.
You can have workaround for fallback intents but there is absolutely no way of getting accuracy score of intents as of now.
You may want to check some other NLP engines like Dialogflow, RASA, LUIS if you want accuracy score of the intents.

sid8491
  • 6,622
  • 6
  • 38
  • 64