LinkedIn
Copied!

Table of Contents

Transferring training data to another email bot

Version:

Only available versions of this content are shown in the dropdown

You can take greater control of the quality of the text analytics model in Pega Email Bot, and in your application, by copying training data from one email bot environment to another environment. Copying training data from one environment to another results in a higher number of triaged emails, greater accuracy of topic and entity detection, and makes the email bot easier to maintain and build.

For example, you can copy training data from a production environment to a development or a QA environment. Before you can begin transferring data, you first start recording training data in the email bot. For more information, see Enabling the training data recording for an email bot.
To prepare training data for an email bot, modify training data in the email bot. For more information, see Correcting training data in an email bot.

Copying training data from one email bot environment to another involves first creating training data in a production environment, and then exporting the training data to a file. Next, you edit another email bot in a development or QA environment and import the saved training data from the file. You can export either only the reviewed training data for an email bot, or all of the training data. The training data that you export includes entities that the system added to the text analyzer but did not build in the model.

  1. Save training data from an email bot, for example, in a production environment.

    For more information, see Exporting training data for an email bot.

  2. Reuse training data in another email bot, for example, in a development or QA environment.

    For more information, see Importing training data for an email bot.

Update the text analytics model for the second email bot by using the copied training data. For more information, see Applying changes to a text analytics model for an email bot.

  • Training data for the Email channel

    To use Pega Email Bot in your application to seamlessly respond to user problems, train the system to recognize different user input in emails, such as help requests or issues. When you train the data for the email bot, the system learns from training records, improves the artificial intelligence algorithms, and provides better responses to user input.

Have a question? Get answers now.

Visit the Collaboration Center to ask questions, engage in discussions, share ideas, and help others.