Skip to content
This repository has been archived by the owner on Jul 2, 2019. It is now read-only.

Use the LUIS score when processing an action #5

Open
dtzar opened this issue Jun 6, 2016 · 1 comment
Open

Use the LUIS score when processing an action #5

dtzar opened this issue Jun 6, 2016 · 1 comment

Comments

@dtzar
Copy link
Member

dtzar commented Jun 6, 2016

From @f2bo on May 27, 2016 17:33

If a user expresses an intent that is not defined by the LUIS model, it can be mistakenly mapped to one of the methods in the ActionDialog. Currently, the implementation assumes that the intent was interpreted correctly, which may lead to unexpected results. For example, see #78.

The LuisResult includes a score that expresses the confidence in the result. There are scores for both intents and entities. We could take advantage of the scores to decide whether to execute the corresponding action or ask for confirmation. For example, if the score is below a certain threshold, reply with a message saying "I didn't understand. Did you mean XXXX?"

Copied from original issue: dtzar/AzureBot#86

@dtzar
Copy link
Member Author

dtzar commented Jun 22, 2016

This might be good considering there are a couple recent interactions which make no sense. For instance this one on Slack:
image
and this one on Groupme:
image

@dtzar dtzar added the bug label Jun 22, 2016
@nzthiago nzthiago self-assigned this Sep 29, 2016
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Projects
None yet
Development

No branches or pull requests

3 participants
@nzthiago @dtzar and others