Virtual agent NLU Model

AbdulH743269785
Tera Contributor

I’ve created an NLU model with an intent for password-related issues, using utterances like:

  1. "Need to change my password for Microsoft Teams"

  2. "Forgot my password for Facebook"

  3. "Help me reset the password for Salesforce"

  4. "How can I reset my password for Slack?"
    ...and similar examples.

However, the issue is that whenever I type just the word "password" or "reset", it incorrectly triggers this intent, even though I don’t want it to.

Additionally, the model is automatically mapping software names like Zoom to the software entity, even when that’s not intended.

0 REPLIES 0