Published Release Notes

Train machine learning models for extracting named entities and detecting intents

Data scientists can train machine learning-based text extraction and intent detection models by using the Analytics Center. With text extraction, you can train a Conditional Random Fields (CRF) model to detect whether the content contains specified entity types such as person names, company and organization names, locations, dates and times, percentages, and monetary amounts. For intent detection, you can train a maximum entropy model to understand user intentions expressed in written content. With these two new capabilities, you can quickly react to customer queries and comments by taking appropriate action against the information that you extracted.

For more information, see Creating machine learning-based text extraction models and Creating machine learning-based intent analysis models.

Setting a build version of the custom mobile app

When you build a custom mobile app, you can now configure a build version and leave the custom mobile app version unchanged. This way you can upload an updated custom mobile app to the Apple App Store and Google Play Store without having to increment the app version.

For more information, see Setting the app build version.

Edit icons and titles for messages in the activity feed

You can now edit the icons and titles of messages that are displayed in the activity feed for Pulse by customizing the pxFeed Pulse gadget. For example, to inform other users of an application about important Twitter updates that are related to the organization, you can create a message that displays a Twitter icon and a title of Twitter updates.

For more information, see Pulse enhancements for social collaboration.

Activity feed displays application messages

The activity feed for Pulse now displays messages within the context of the application of the logged-in user. If a user posts messages within the context of the application, all the users of the application can view the messages in their activity feed. Users can select the filter for application messages to view only these messages. For example, application messages can be used for release announcements.

For more information, see:

• Pega^® Express 7.4 – Communicating with Pulse
• Pega Platform 7.4 – Pulse enhancements for social collaboration

pxFeed replaces pxContextFeed and pxActivityFeed Pulse gadgets

The pxContextFeed and pxActivityFeed gadgets have been replaced by the pxFeed gadget to integrate the functionalities of viewing and posting messages in the Pulse activity feed. You can include this gadget within a section or harness across your application, for example, in the dashboard of the Case Manager portal, to display the activity feed. You must add the latest UI Kit ruleset to ensure that your application uses the pxFeed gadget instead of the pxContextFeed and pxActivityFeed gadgets to generate the activity feed.

For more information, see Pulse gadget and custom APIs.

New shape to preload a data page in a case

You can now use the Load data page shape to preload a data page before it displays information to a user in a case. By preloading a data page, you can more quickly display information to a user.

For more information, see Preloading a data page.

Support for NLP model training in the Facebook and WebChat user channels

You can now train and enhance the natural language processing (NLP) model in a Facebook or WebChat user channel to support machine learning. You can review user statements and questions, edit the text, and map content to specific text analyzer categories.

For more information, see NLP model for a Facebook channel and NLP model for a WebChat channel.

Label changes for text analytic models

The classification analysis label has changed to topic detection and the entity extraction label has changed to text extraction. Also, the sentiment analysis, topic detection, and intent detection labels are now located under Text Categorization in the Analytics Center. These name changes reflect industry standards and provide a clearer distinction between different types of text analytics models in Pega^® Platform.

For more information, see Text analytics models.

Automatically detect the most important themes in text

Text analyzers can now automatically recognize the most important concepts that are expressed in text and mark such concepts as entities of type auto_tags. You can use auto_tags to group similar content by its themes and reduce the dimensionality of text to the most important features.

For more information, see Text extraction analysis.

Data pages can source information from a robotic process automation

When you use robotic process automation (RPA), you can now configure data pages to source information from robotic automations so that you can connect your Pega® Platform application to legacy applications in your enterprise. By using automations to retrieve data and save it to a data page, you can use data virtualization to encapsulate your Pega Platform data model from the physical interface of a legacy system against which the automation is running.

For more information, see Obtaining information from robotic automations.