Correcting identified topics

If you want to ensure that Pega Email Bot is accurately detecting the topic and intent of the emails it receives, you can review and correct the topics in the training data records. You train the system by correcting the identified topics in each training record, and then rebuilding the text analytics model with the updated information. This improves the accuracy of the cases and responses that the email bot suggests when it detects the relevant topic. For example, when a user sends an email requesting a car insurance quote, the email bot detects car insurance as the topic and suggests the Car Insurance case during the email triage stage.
Before you begin: Enable recording of training data. For more information, see Enabling the training data recording for an email bot.
  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 Email channel.
  3. In the Email channel, click the Training data tab.
  4. Optional: If you configure multiple languages for the email bot, to filter data records by a language, in the Language list, select a language.
    For example: To display data records only detected in the French language, select French.
  5. In the list of training records, select a data record.
    Result: The NLP analysis section displays the detected topic for the selected training data record.
  6. In the Topic field in the NLP analysis section, press the Down arrow key, and then select a more appropriate topic for the training record.
    For example: To correct a training record so that its intent relates to car insurance, select Car insurance.
  7. Optional: To use this training record to improve the artificial intelligence of the email bot, in the Review training data section, click Mark reviewed.
    Create at least 15 records in the training sample to improve detection of the right information in emails.
  8. Optional: To correct the identified topics in additional training records, repeat steps 4 through 7.
  9. Click Save.
What to do next: Teach the email bot the reviewed and corrected training records by rebuilding the text analytics model. For more information, see Applying changes to a text analytics model for an email bot.