Importing training data for an IVA

You can copy training data from one Pega Intelligent Virtual Assistant (IVA) to another IVA, to ensure the correct detection of topics and entities. Using training data with your IVA improves the text analysis and artificial intelligence of the system, so that the IVA responds correctly and as expected during chat conversations with users. You copy the training data between IVAs by exporting the data to a file, and then importing the file to the next IVA. For example, to use training data in an IVA in a QA environment, you export the data from an IVA that already detects the right information in a production environment. You then import the data from the file to the IVA in the QA environment.
Before you begin: Export training data from an IVA to a .zip file. For more information, see Exporting training data for an IVA.
  1. In the header of Dev Studio, click the name of the application, and then click Channels and interfaces.
  2. In the Current channel interfaces section, click the icon that represents your existing Unified Messaging or Web Chatbot channel.
  3. In the channel, click the Training data tab.
  4. Click the More icon above the list of training records, and then click Import data.
  5. In the Import data window, click Choose file, and then select the .zip file that contains the exported training data.
  6. Click Import.
What to do next: Update the text analytics model for the IVA that contains the imported training data. For more information, see Applying changes to a text analytics model for an IVA.