Sebastian Glock
Authors name: Sebastian Glock February 15, 2022

Many of our customers have utilized the Intent feedback analyzer to continuously improve their NLU models with tremendous success. However, at Cognigy, we are regularly asked about a more systematic approach to Intent training in projects.

So, we created a comprehenisve tutorial to present a systematic technique that can be employed during the project planning process.

Visit our Help Center Article to delve deep into a systematic approach to Intent training.


In the article, you'll learn about the phases that can be used to outline the intent creation/training process:


Improving Intent recognition

  • Machine learning

  • Cognigy Script

Resolving Intent conflicts

  • Moving Example Sentences to the better-suited Intent

  • Moving Intents into a hierarchy

  • Merging multiple Intents into one

  • Outsourcing Intents to a separate Flow

  • Adjusting the NLU settings

Reducing false positives

  • Machine learning / Reject Intent

  • Adjusting the NLU settings