How to use machine learning models
Once your prediction model is created, set up the actual predicting as a business process using the [Predict data] process element. This gives you complete control as to what records are predicted and when.
For example, you can set up prediction of account category whenever a new account with empty [Category] field is saved (Fig. 1).
In this example, we will be using the account category prediction model created earlier.
To implement this:
1.Create a new business process and add the [Signal] start event on its diagram. The start event should be triggered whenever a new record in the [Accounts] section is added. Signal element parameters (Fig. 2):
a.[Which type of signal is received?] - “Object signal”.
b.[Object] - “Account”.
c.[Which event should trigger the signal?] - “Record added”.
d.[The added record must meet filter conditions] – “Category is not filled in”.
e.[Run following elements in the background] – “true”. This way, all preceding system operations of the process will be performed in the background without displaying the loading mask.
2.Add the [Predict data] element on the diagram. Set up the element parameters (Fig. 3):
a.In the [Machine learning model] field, choose the prediction model to use. For example, to predict the category of the account, select the “Account category” model created earlier. The model setup is described in the “How to create a lookup value prediction model” article.
Note
Prediction models have to be trained before they can be used in the business process. If the model is not trained, it will not be available for selection in the [Predict data] element.
b.[What type of prediction to use?] - “Predicting for one record”.
c.In the [What record to perform prediction on?] field, click the button and select [Process parameter]. In the window that appears, go to the [Process elements] tab and select the signal created in the previous step, and then select [Unique identifier of record].
3.Save the process.
As a result, whenever the [Predict data] element is triggered during a business process, it will use the specified ML model to predict the data of the specified record. In our case, the [Category] field value will be predicted and populated each time a new record is saved in the [Accounts] section. The prediction will be based on the values specified by users when populating the [Category] field of historical records.
See also
•Basic predictive analysis glossary
•How to create a lookup value prediction model
•How to create a numeric value prediction model