Configure enhanced natural language user input handling

Add enhanced natural language handling for boolean user responses to your Virtual Agent environment. If you're using additional languages or third-party NLU providers, specify the model and intents that define variations in yes/no utterances for your bot.

Before you begin

In your NLU provider, create a model with two intents, or add two intents to an existing model. One intent should contain affirmative or "yes" utterances. The other intent should contain negative or "no" utterances.

If your environment supports multiple languages, create models and intents for each language.

Role required: virtual_agent_admin or admin

About this task

The default entries in this table apply to ServiceNow NLU for English only. If you want to implement natural language handling in other languages, add new entries to the table for each language.

Procedure

1. Navigate to All, and then enter sys_cs_nlu_dispatch.list in the filter.
2. Select New or an existing entry that you want to modify.
   Alternatively, select New to add an entry for another language or third-party NLU provider.
3. On the form, fill in the fields.

   Field	Description
   NLU Model ID	Name of the model that contains the defining intents for the boolean responses.
   Name	Value name that represents the type of user input being configured. The following values are valid: BOOLEAN_NO: Maps to utterances that express a "no" response to a yes/no question. BOOLEAN_YES: Maps to utterances that express a "yes" response to a yes/no question.
   Active	Option that enables enhanced natural language user responses on your instance.
   NLU Intent ID	Name of the corresponding intent for the boolean option. By default, the intent for BOOLEAN_NO is Negative. The intent for BOOLEAN_YES is Affirmative. If you need to make changes, create a new model with these intents, and then specify the new intent in this field.
   Language	Language of the intent. English for ServiceNow NLU is configured by default. If you need to support other languages,you can add a row in the table for each supported language.

4. Select Submit or Update.

Result

Depending on your environment, your configuration may contain entries for other languages (such as French or German in the example below) or for other NLU providers, such as IBM Watson.