Defining conversational channel behavior

Improve user experience in Pega Intelligent Virtual Assistant (IVA) by configuring the conversational channel behavior that fit the needs of your organization. You can add a conversation to a case type, add suggested cases, and define suggested responses for the IVA so that you speed up business processes for a Pega Platform application. You can also configure text analyzer settings and system responses to improve the artificial intelligence of the IVA.
Before you begin: Create a conversational channel to use as an IVA. For more information, see Creating a conversational channel.
  1. Prepare the IVA to collect case information and create a case with the information:
    1. Configure the IVA to automatically collect case information through standard questions to users, by adding a conversation to a case type.
      For more information, see Adding a conversation to a case type.
    2. Configure the system to create a case in a Pega Platform application based on user input, by adding case commands in the IVA.
      For more information, see Adding case commands for a conversational channel.
  2. Configure the system to provide automatic responses to user actions by defining response commands for the IVA.
    For example: The system can authenticate, display text messages and a menu of options, or perform other actions as a response to user input.
    For more information, see Defining response commands for a conversational channel.
  3. Configure the IVA to provide the best response to user input by detecting topics and entities:
    1. Define topics, which are the general subject and intent of user input that is detected by the IVA using text analysis.
      For more information, see Defining topics for text analysis for an IVA.
    2. Define entities, which are the proper nouns that fall into a common category, that are detected by the IVA using text analysis.
      For more information, see Creating entities for an IVA.
    The IVA sets up the iNLP text analyzer by default. The iNLP text analyzer uses natural language processing (NLP) and adaptive analytics text analysis in interactions with users to detect the topics and entities that then determine the best matched response. For more information, see Configuring text analyzer settings and Exploring text analyzers.
What to do next: Perform the following tasks: