Train predictive models

Apromore now allows to train predictive machine learning models from an event log, analyze the quality of these models, and generate predictions.

Given an event log, we can train a machine learning model to predict:

  1. The remaining time of an ongoing case until its completion.

  2. Whether an ongoing case will reach a given outcome or not, for example, will a case in a price quotation process reach a state where the customer places an order or will an order-to-cash process reach a state where the customer returns the product.

To use the plugin, first, select the log and right click on Monitor -> Manage predictor.



As an alternative, select the log and right click on it, choose Manage predictor -> Create new predictor.


Predictor Trainer window opens, containing 3 tabs: Train - Validate - Explain. Train is opened by default.


Type the The Predictor name and select the Prediction type and Target attribute in the corresponding fields.


To start the predictive models training it is mandatory to fill only The Predictor name and Prediction type.


Input attributes will appear, based on the uploaded log attributes. By default they all selected. To unselect/select back the attribute – untick/tick the box next to the attribute name.


Click Train predictor. The pop-up message stating the status of the training will appear. The status is also displayed in the upper right corner of the Predictor Trainer window right beneath the user details.



Active status written in green means that the predictor is ready to be used. Failed status written in red means that some error occurred during the training. The log should be rechecked. Training status means that the predictor is getting prepared to be used. Depending on the size of the log, it may take some time to generate the predictor.


To check the accuracy of the predictor, switch to the Validate tab.


Change the Validation type by clicking on the corresponding arrow and choosing from the drop-down list of available types.


To analyze the logic behind the predictor deeply and check which attributes were used for the predictions, switch to the Explain tab.