Applying simulation changes to a text analytics model
After you make corrections to natural language processing (NLP) records from the preview console relating to topic, entities and language for Pega Intelligent Virtual Assistant™ (IVA), apply the changes to the text analytics model. Rebuilding the text analytics model with additional or modified records improves the IVA's ability to understand user requests, and improves how the system responds to detected topic, entities, and language in the user input.
When you update a detected topic or add entities during a chat simulation in the preview console, the system saves the changes as a training record.- Test the chatbot responses. For more information, see Verifying detected topics and entities.
- Update the chatbot questions. For more information, see Modifying chatbot response questions.
- Obtain data from a conversation. For more information, see Creating entities for an IVA and Verifying detected topics and entities.
- In the navigation pane of App Studio, click Channels.
- In the Current channel interfaces section, click the icon that represents your existing Digital Messaging or Alexa channel.
- In the channel, click the Behavior tab.
Result: The number of records that you previously modified is displayed at the top of the preview console. - At the top of the preview console, click Build
model.
For example: The following figure shows where in the preview console you build the text analytics model for the chatbot: The preview console used to simulate and built an IVA
Previous topic Mapping entities in conversation text Next topic Training the model for the IVA channel