Published Release Notes

Update text analysis models during every interaction

While interacting with your customers through text-based communication channels (such as a Facebook chatbot or email), you can use an activity to provide feedback to a text analysis model by manually assessing whether the entities, categories, intents, and sentiments in the analyzed text were assigned to the expected class. By using this feedback input as training data, you can ensure that the accuracy of your text analytics models continuously improves.

For more information, see Feedback loop for text analysis.

Build predictive models with the genetic algorithm

A genetic algorithm solves optimization problems by creating a generation of possible solutions to the problem. In Pega^® Platform you can utilize the functionality of genetic algorithms by using the genetic algorithm model in the Analytics Center portal. With this type of predictive model, you can create highly predictive, non-linear models by selecting the best performing model from the last generation.

For more information, see Creating models and Creating a genetic algorithm model.

The Run Data Flow shape replaces the Run Interaction shape

The Run Interaction shape in flows has been replaced by the Run Data Flow shape to introduce a consistent way of making decisions and capturing responses. You can now run a single-case data flow by referencing the data flow in the Run Data Flow shape. The data flow must have an abstract input and abstract or dataset output.

For more information, see Running a decision strategy from a flow, About Data Flow rules and Interactions in flows are no longer supported by the Run Interaction shape.

Interactions in flows are no longer supported by the Run Interaction shape

The Run Interaction shape in flows has been replaced by the Run Data Flow shape, which supports running a single case data flow with a strategy. Flows that include the Run Interaction shape continue to work; however, you must now use the Utility shape to reference any new interactions that you create.

For more information, see Running a decision strategy from a flow and About Interaction rules.

Specify a model transparency policy for predictive models

Specify a model transparency policy in Pega^® Platform to indicate predictive models that are compliant or non-compliant to use in particular business issues. Increase the transparency threshold of particular business issues to allow the use of more transparent models. Decrease the transparency threshold to allow for the use of less transparent and opaque models. Models with a transparency score below the transparency threshold are marked as non-compliant.

For more information, see Model transparency for predictive models.

Define custom dimensions in VBD data sets

Beginning with version 7.3.1, you can define custom dimensions when you create a Visual Business Director (VBD) data set in Pega^® Platform. With this option, you can use any custom or existing property from the Applies To class of the VBD data set as a dimension to visualize your data, in addition to the properties in the default set of dimensions.

For more information, see Creating a Visual Business Director data set record.

Extension attributes are not supported in PMML models

Models in the Predictive Model Markup Language (PMML) format version 4.3 that contain extension attributes with the x- prefix are not valid. These extension attributes are deprecated; you must use extension elements instead. In addition, if the output type of any output field in the model is set to FLOAT, change it to DOUBLE.

For more information, see PMML 4.3 - General Structure in the Data Mining Group documentation.

The Upload responses action is not supported for adaptive models with customized context

A default instance of the Adaptive Model rule contains five model identifiers (.pyIssue, .pyGroup, .pyName, .pyDirection, .pyChannel) that are used to partition adaptive models. If you add other identifiers in your Adaptive Model rule instance, you cannot upload responses to this instance with the Upload Responses wizard and the following error is displayed: The Flow Action post-processing activity pzUploadCSVFile failed: Cannot parse csv file.You can still train such adaptive models with data flows.

For more information, see Training adaptive models in bulk with data flows, Model context, and Uploading customer responses.

Decision Analytics portal renamed to Analytics Center

The Decision Analytics portal is renamed to the Analytics Center portal. In this work area for predictive analytics and text analytics, the business scientist can control the full model life cycle by building predictive models, importing PMML models, building sentiment analysis and text classification models, creating and monitoring adaptive models, and updating any existing models.

For more information, Analytics Center portal.

Upgrading Adaptive Decision Manager data mart tables might fail

Issue: Upgrade from 7.3 to 7.3.1 fails if the data contained in the pxInsName column of the PR_DATA_DM_ADMMART_PRED_FACT table is longer than 128 characters.

Reason: During the Pega Platform™ upgrade from 7.3 to 7.3.1, data in the Adaptive Decision Manager (ADM) data mart tables is migrated from the PR_DATA_DM_ADMMART_PRED_FACT table to the PR_DATA_DM_ADMMART_MDL_FACT table. In Pega 7.3.1, ADM uses only the PR_DATA_DM_ADMMART_MDL_FACT table where the pxInsName property can store values that are 128 characters long. In Pega Platform 7.3, the pxInsName property in the PR_DATA_DM_ADMMART_PRED_FACT table can store values that are 255 characters long. If the pxInsName property contains values that are longer that 128 characters, the upgrade fails.

Resolution: Issue an ALTER TABLE statement to change the pxInsName column size to 255 characters and resume the upgrade. For example:

ALTER TABLE rules.pr_data_dm_admmart_pred ALTER COLUMN pxInsName TYPE varchar(255);

For more information, see Adaptive Decision Manager data model.