INC-183211 · Issue 667012
Compare-To-Empty-String support added to Analyzer
Resolved in Pega Version 8.7
An enhancement has been added to support a NotEquals decision function in the Analyzer.
INC-183431 · Issue 665162
Strategy execution queries tuned
Resolved in Pega Version 8.7
During NBA strategy execution, interaction history summaries were queried multiple times for a single strategy execution. To improve this function, the isChanged query will now be executed once after system startup if there are no changes instead of being executed every time the Aggregates DataSet BBK is executed. If the same Aggregates query is made multiple times as part of a Strategy in a REST request (i.e. likely multiple ADM models in a MakeDecision call), only one Aggregates BBK call will be made.
INC-183855 · Issue 667692
Feedback data visible after update
Resolved in Pega Version 8.7
In order to improve backwards compatibility, handling has been added for blank values of ruleset and classname in feedback data from NLP models.
INC-184352 · Issue 671897
Predictive model snapshot generation updated for Java 11
Resolved in Pega Version 8.7
While the Simulation data export case type was executed as expected, the predictive stages showed zero records processed. This was traced to instances running on Java 11 which caused an issue with the snapshot process. This did not affect learning, but visualization and reporting were not working in Prediction studio. To resolve this, an update has been made to the snapshotID for Java 11 compatibility which will generate a random UUID instead of using a seed.
INC-185045 · Issue 672416
Performance improvements for action check-in
Resolved in Pega Version 8.7
The process of 'Saving As' an action to a higher ruleset, Saving, and then running Check-in was taking an excessive amount of time. This was traced to redundant page copy activities and performance has been improved by updating the implementation.
INC-186437 · Issue 685015
Updated entity attachment extraction tokenizers
Resolved in Pega Version 8.7
After creating an entity extraction model, it was seen that one of the entities worked when there was a space after the semicolon but the detection was not working if there was no space. This has been resolved by updating the Tokenizers with extra examples to address tokenization when ":" is present between two words without any spaces.
INC-186889 · Issue 674014
Unused DSS compiledExpressions now defaults to false
Resolved in Pega Version 8.7
After updating from Pega 8.2 to Pega 8.5, executing a test campaign failed with an error in logs. This occurred when running Distribution tests with "Include funnel analysis reports" checked, and was traced to the DSS setting "decision/strategy/ssaVM/compiledExpressions". As this was previously used to toggle a feature which is no longer in use, the default setting for this DSS has now been changed to false to resolve this issue.
INC-187070 · Issue 672106
Strategy execution queries tuned
Resolved in Pega Version 8.7
During NBA strategy execution, interaction history summaries were queried multiple times for a single strategy execution. To improve this function, the isChanged query will now be executed once after system startup if there are no changes instead of being executed every time the Aggregates DataSet BBK is executed. If the same Aggregates query is made multiple times as part of a Strategy in a REST request (i.e. likely multiple ADM models in a MakeDecision call), only one Aggregates BBK call will be made.
INC-187520 · Issue 677873
Actuals sync clears data from cache and persistence
Resolved in Pega Version 8.7
When sync attempted to clear the existing data from Actuals before copying data from the interaction history, the partition was cleared from the cache but not from persistence. IH data that was then copied to the same partition time ranges had new field descriptors created that did not match the format of the old data in persistence. As a result, when a query triggered the loading of such a partition, the mix of data with the old and new formats caused the load to fail. This has been resolved by updating the system to ensure the data is deleted properly from both the cache and persistence.
INC-188124 · Issue 672587
Multistage Prediction Logic improved
Resolved in Pega Version 8.7
When Predictions with Response wait time configured were executed as part of make decision, each decision that came out of the strategy caused a pxDR call to fetch the relevant model execution results and process them to stage the default outcome results for models. This has been improved by modifying the system to process the existing records rather than reading data again from pxDR.