Published Release Notes

Support for custom database tables in external Cassandra clusters

Pega Platform™ now supports a connection to external Cassandra clusters through a dedicated Database Table data set, which reduces the need for data ingestion and export. You can use custom tables that you store in your external Cassandra cluster in data flows for accessing and saving data. You can access your custom data model by mapping the model to a Pega Platform class.

For more information, see Connecting to an external Cassandra database through a Database Table data set.

Text analytics models editing and versioning

Pega Platform™ now supports editing and updating training data for text analytics models.

Pega Platform also supports the versioning of text analytics models. When you update the model, Prediction Studio creates an updated model version. You can then switch between the model versions.

Upgrade impact

In versions of Pega Platform earlier than 8.3, the training data for text models was stored in the database. In Pega Platform version 8.3 and later, the training data for text models is stored in Pega Repository. You cannot build new models without setting the repository. After the repository is set, all text models are automatically upgraded and will work normally.

What steps are required to update the application to be compatible with this change?

After a successful upgrade, set the repository in Prediction Studio before building or updating any Natural Language Processing (NLP) models.  In Prediction Studio, click Settings > Text Model Data Repository.

For more information, see:

• Increase the accuracy of text analytics models by adding feedback data (8.3)
• Updating training data for text analytics models

Text analytics models migration

Pega Platform™ now supports the exporting and importing of text analytics models. For example, you can export a model to a production system so that it can gather feedback data. You can then update the model with the collected feedback data to increase the model's accuracy.

Upgrade impact

In versions of Pega Platform earlier than 8.3, the training data for text models was stored in the database. In Pega Platform version 8.3 and later, the training data for text models is stored in Pega Repository. You cannot build new models without setting the repository. After the repository is set, all text models are automatically upgraded and will work normally.

What steps are required to update the application to be compatible with this change?

After a successful upgrade, set the repository in Prediction Studio before building or updating any Natural Language Processing (NLP) models.  In Prediction Studio, click Settings > Text Model Data Repository.

For more information, see:

• Increase the accuracy of text analytics models by adding feedback data (8.3)
• Exporting text analytics models
• Importing text analytics models

Prediction Studio notifications

Prediction Studio now displays notifications when adaptive or predictive models encounter problems. For example, if the performance of a model decreases suddenly, a notification appears, which you can use to immediately investigate the cause of the problem.

For more information, see Accessing Prediction Studio notifications and Prediction Studio notification types.

Production data sampling and migration for decision strategy simulations

Pega Platform™ now supports the sampling and migration of data from the production environment to the simulation environment. The data that is migrated for simulations includes customer details, Adaptive Decision Management information, and interaction history. By running a simulation on that sample data in Pega Marketing™ or Pega Customer Decision Hub™, you can verify how the changes to your decision logic impact the results. You can use that information to optimize and adjust your decision algorithms and processes.

For more information, see Deploying sample production data to a simulation environment for testing.

Integrate text analytics with decision strategies through the Interaction API

The Interaction API provides more context for making next-best-action decisions by integrating text analytics with decision management components such as strategies, propositions, and interaction history. By including natural language processing in your decisioning solution through the Interaction API, you ensure that the next-best-action decisions that you make are more informed and accurate.

For more information, see Customizable Interaction API for text analytics.

Support for machine learning as a service models

In Pega Platform™, you can now run advanced machine learning and artificial intelligence models that you develop in third-party tools. By configuring a connection with an external machine learning as a service provider, such as Google AI Platform, you can use the predictive power of your custom models to improve the predictions in your customer strategies.

For more information, see Connecting to a Machine Learning as a Service model and Configuring a machine learning service connection.

Improved management of batch indexing

You can now cancel and check the status of a batch index process directly from the Search landing page in Dev Studio. The landing page now refreshes automatically every 10 seconds so that you can easily see the most recent status of the batch index process. Additionally, the reindex operation is now 20% faster than in previous versions. These features provide greater visibility into batch indexing and improve your ability to fix issues with a batch indexing process.

For more information about batch indexing, see Rebuilding search indexes from the user interface.

.

Default filter settings for Pulse feed sources

Set default filters for Pulse messages, so that users see only relevant posts. Users can temporarily change these settings at run time to view messages from different feed sources, and restore the default settings by refreshing the page.

For more information, see Adding a feed source for the activity feed.    

Stream-based alternative for interaction history

Accelerate the processing of high-volume interaction history data by incorporating a stream-based interaction history into your application. A stream-based interaction history processes large volumes of data more quickly than a traditional relational database interaction history so that you can react to your customers' needs in real time.

For more information, see Process high-volume interactions more efficiently and Aggregates-only mode for a stream-based interaction history.