Creating predictions for case management
Optimize case automation by creating predictions that predict case events. You can create predictions to predict a business outcome and use this information to route cases or prioritize work.
For example, a prediction that predicts the probability of
fraud can help you route suspicious cases for closer inspection. By predicting the
probability of successful case completion, you can prioritize cases that are most likely
to bring you value or prioritize work that needs immediate attention.
What to do next: Add the prediction to a case type. For more information,
see Predicting case outcomes.
- In the navigation pane of Prediction Studio, click Predictions.
- In the header of the Predictions work area, click New.
- In the Create a prediction window, select Case management, and then click Next.
- In the Prediction name field, enter a name for the prediction.
- Specify what you want to predict.
For example: To predict whether a case is likely to be completed successfully, in the Outcome list, select Case completion.
To predict an outcome that does not match any of the available templates, select the following settings:
- In the Outcome list, select Custom.
- In the Outcome name field, enter a name for the outcome, for example, Future customer lifetime value.
- Define the outcome type:
- To predict whether a future event will happen or not (binary outcome), select Two categories.
- To predict a numerical value (regression modeling), select Continuous value.
Creating a prediction to predict the probability of fraud 
- In the Subject list, select the case type or data type
with which you want to associate this prediction.For a case completion prediction, you can only select a case type as the subject. For a claims fraud or custom prediction, you can only select data types.
- If you work with branches and want to save the prediction to a specific branch, select Save to branch, and then select a rule set and a branch.
- Review your settings, and then perform one of the following actions:
- If you selected Claims fraud or a similar template as the outcome that you want to predict, click Create and go to step 15.
- If you selected the Case completion template, start the Prediction wizard by clicking Start wizard.
- In the Select data wizard step, select I do not
have historical data.
Note: As a best practice, do not use historical data to pretrain a case prediction because historical case data may have been overwritten as the case progressed. You might still use the I have historical data option to gain insights into the available predictor information. - Click Next.
- In the Prediction configuration wizard step, review the response labels for the prediction, and then click Next.
- In the Select predictors wizard step, select the fields
that you want to use as input for the prediction.To increase the accuracy of your prediction, select a wide range of fields to use as predictors. Do not include fields that are not suitable as predictors, for example, the Identifier and Date Time fields. For more information, see Best practices for choosing predictors.
- Confirm your choice of predictors by clicking Next.
- In the Review prediction wizard step, review the prediction settings, and then complete the model creation process by clicking Create.
- To change the prediction properties at this stage, in the Prediction window, click Configure, and then make your changes.
- Click Save.
Result: The prediction is now available in the Predictions workspace. Depending on the settings that you selected, Prediction Studio created an adaptive model or a scorecard as the basis of the prediction. - If you are using parameterized predictors, add them to the adaptive model that
is the basis of the prediction.For more information, see Adding parameterized predictors.
Tip: To open the model that is the basis of the prediction, in the Prediction workspace, on the Models tab, click the name of the corresponding model. - Optional: To replace the model that is the basis of the prediction with a different model, see Replacing models in predictions with MLOps.
Previous topic Creating and managing predictions Next topic Creating text predictions