Activating and training adaptive models for artificial intelligence in Pega Sales Automation for Financial Services (8.4 and prior)
Artificial intelligence insights predict both the probability of and the date when an opportunity moves to the next stage or closes. Predictions based on artificial intelligence can help sales representatives to proactively assess risks on deals in the pipeline.
Before using artificial intelligence insights with the Pega Sales Automation for Financial Services™ application, configure the application to train Pega’s adaptive models for artificial intelligence. Log in to Pega Sales Automation for Financial Services and complete the following steps:
Activating opportunity insights
- In the Dev Studio header search text field, search for and select the EnableOpportunityInsights dynamic system setting.
- On the Settings tab, in the Value field, enter
true
. - Click .
- Click Status of Normal.
- Decision Data Store
- Adaptive Decision Manager
- Data Flow
- Visual Business Director
and verify that each of the following services contains a node with a
Activating the failsafe mechanism
- Click Configure > Decisioning > Infrastructure > Services > Adaptive Decision Manager.
- In the Adaptive decision manager nodes section, click .
- On the Adaptive decision manager form, select the Enable failsafe mechanism check box and click Submit.
Training steps
The data rules training is available in the Pega Sales Automation for Financial Services sample application rulesets. For more information, see Sample data flows in Pega Sales Automation for Financial Services.
For B2B sample data training, first run the activity:
- In Dev Studio, search for PopulateB2BOppInsights and select the PegaSAFS-Data-Opp-Predictors activity.
- Click Actions > Run. The process takes approximately 10 minutes to complete.
Next, to track the data flows, perform the following steps:
- Click Configure > Decisioning > Decisions > Data Flows > Batch Processing.
- Review the presented configuration.
For B2C sample data training, first run the activity:
- In Dev Studio, search for PopulateB2COppInsights and select the PegaSAFS-Data-Opp-Predictors activity.
- Click Actions > Run. The process takes approximately 10 minutes to complete.
Next, to track the data flows, perform the following steps:
- Click Configure > Decisioning > Decisions > Data Flows > Batch Processing.
- Review the presented configuration.
Re-training steps
After the training steps are performed, if you want to retrain the data, clear the strategy results or already captured model responses:
- In Dev Studio, search for the pxDecisionResults data set.
- Click Actions > Run.
- From the Operation field, select Truncate and then click .
- In Dev Studio, search for the PreviousStages data set.
- Click Actions > Run.
- From the Operation field, select Truncate and then click .
- Click Configure > Decisioning > Adaptive Model Management and select all of the available PegaSAFS-Work-Opportunity and PegaSAFS-Work-Opportunity-Ind models.
- Click .
- Optional: To capture new opportunity insights, perform the training steps, if necessary.
Previous topic Deprecated and withdrawn rules and table changes in Pega Sales Automation for Financial Services Next topic Pega Sales Automation for Financial Services modified rules for BAC prevention (8.3 and prior)