Published Release Notes

Real-time event processing

Real-time event processing is a mechanism that can be used to analyze a high velocity stream of events entering the system and derive conclusions from them. It is possible to define metrics, measurements, and conditions under which subsequent trigger events are generated.

For details, see Real-time event processing.

Text analytics on the Pega 7 Platform

The Pega 7 Platform offers users the capability to analyze units of text, extract quality information, and translate it into business intelligence. The text analytics functionality aims to determine the positive, negative, and neutral polarities (tone) of subjective sentences. It also provides mechanisms for categorizing units of text and classifying text elements into predefined categories. You can use it to analyze text-based content including news feeds, emails, and postings on social media streams, such as Facebook, Twitter, and YouTube.

For more information, see Introduction to text analytics on the Pega 7 Platform.

Updates to the Data Flow rule

It is now possible to use report definitions as primary and secondary sources in the Data Flow rule. The data flows that are sourced by a report definition can be distributed on multiple nodes. When using a database table data set as the destination of the data flow, you need to select one of the save options for the data set (insert records or insert and update records).

Updates to the Event Strategy rule

When you design event strategies, you can use the Filter shape for basic arithmetic and text expressions. You can also specify start conditions for the tumbling and sliding windows.

Updates to the Predictive Model rule form

The Predictive Model rule form provides the XML schema preview for uploaded PMML models. You can use this preview to view the structure of a model and to correct errors before saving the rule. When the model contains errors, they are displayed in the Errors section at the bottom of the rule.

Updates to the Strategy rule

The Strategy tab of the Strategy rule form contains a new Test runs panel from which you can run simulations and test strategies for single and multiple customers. When you test strategies for multiple customers, you can identify the most popular propositions offered to the customers, check the overall strategy performance, and examine the performance of the individual components.

New types of data sets

There are three new types of data sets:

• Facebook - Required for connecting with the Facebook API
• Twitter - Required for connecting with the Twitter API
• YouTube - Required for connecting with the YouTube Data API

Use these data sets when you create instances of the Free Text Model rule and you want to analyze the text-based content on social media streams such as Facebook, Twitter, or YouTube.

Problem with truncating Decision Data Store data set

The Truncate operation for the Decision Data Store data set may cause timeout exceptions. This problem is caused by the Apache Cassandra database that waits until compaction tasks finish before it can truncate the data set.

Recommendation:

Repeat the Truncate operation until it is successful.

Predictive models monitoring

In Prediction Studio, you can now monitor the predictive performance of your models to validate that they make accurate predictions. Based on that information, you can re-create or adjust the models to provide better business results, such as higher accept rates or decreased customer churn.

For more information, see Monitoring predictive models.

Kafka custom serializer

In Kafka data sets, you can now create and receive messages in your custom formats, as well as in the default JSON format. To use custom logic and formats for serializing and deserializing ClipboardPage objects, create and implement a Java class. When you create a Kafka data set, you can choose to apply JSON or your custom format that uses a PegaSerde implementation.

For more information, see Creating a Kafka data set and Kafka custom serializer/deserialized implementation.