Create Predicted Eventlog - Revision history

MarHink at 14:15, 20 January 2026

2026-01-20T14:15:37Z

← Older revision		Revision as of 14:15, 20 January 2026
Line 1:		Line 1:
	This article has instructions how to install, configure and use eventlog predictions. The prediction creates a new model that contains the source model data and the predictions. It's able to predict case attributes for the generated new cases and event attributes for the predicted events. To distinguish the real (source data) and predicted events and cases, there are following attributes in the model:		This article has instructions how to install, configure and use eventlog predictions. The prediction creates a new model that contains the source model data and the predictions. It's able to predict case attributes for the generated new cases and event attributes for the predicted events. By default, predictions use [https://en.wikipedia.org/wiki/Transformer_(deep_learning) Transformer] neural network architecture. However, also [https://en.wikipedia.org/wiki/Long_short-term_memory LSTM] and [https://en.wikipedia.org/wiki/Gated_recurrent_unit GRU]-based architectures are supported.

			To distinguish the real (source data) and predicted events and cases, there are following attributes in the model:
	* Event attribute '''Predicted''' denotes whether the event is from the source data (''false'') or whether it's predicted (''true'').		* Event attribute '''Predicted''' denotes whether the event is from the source data (''false'') or whether it's predicted (''true'').
	* Case attribute '''Generated''' denotes whether the case is in the source data (''false'') or whether the prediction generated it as a new case (''true'').		* Case attribute '''Generated''' denotes whether the case is in the source data (''false'') or whether the prediction generated it as a new case (''true'').

MarHink: /* Attribute configuration */

2026-01-15T11:28:22Z

Attribute configuration

← Older revision		Revision as of 11:28, 15 January 2026
Line 171:		Line 171:
	**** Depending on the context, the default value is any one of the following configurations:		**** Depending on the context, the default value is any one of the following configurations:
	***** ignore_values_threshold_for_case_attribute_values		***** ignore_values_threshold_for_case_attribute_values
			****** Used when clustering case attributes when generating attribute values for generated new cases.
			****** Default value is 0.01.
	***** ignore_values_threshold_for_case_attributes		***** ignore_values_threshold_for_case_attributes
			****** Used when clustering case attributes.
			****** Default value is 0.1.
	***** ignore_values_threshold_for_event_attributes		***** ignore_values_threshold_for_event_attributes
			****** Used when clustering event attributes.
			****** Default value is 0.1.
	* '''columns''': An array of attribute column configuration objects used to define columns in the input data that are to be used as event- or case attributes.		* '''columns''': An array of attribute column configuration objects used to define columns in the input data that are to be used as event- or case attributes.
	** If null, all the columns will be included as categorical attributes (except case id, event type (only for event) and timestamp (only for event) columns).		** If null, all the columns will be included as categorical attributes (except case id, event type (only for event) and timestamp (only for event) columns).

MarHink: /* Attribute configuration */

2026-01-15T11:05:58Z

Attribute configuration

← Older revision		Revision as of 11:05, 15 January 2026
Line 168:		Line 168:
	**** Default value: 20		**** Default value: 20
	**** NOTE: Clustering is used by default to convert a set of attribute values into an input- or output vector used by the prediction model.		**** NOTE: Clustering is used by default to convert a set of attribute values into an input- or output vector used by the prediction model.
			*** '''ignore_values_threshold:''' The minimum percentage of objects having a specific attribute value in order for that attribute value to be taken into account as unique attribute value within this categorical group.
			**** Depending on the context, the default value is any one of the following configurations:
			***** ignore_values_threshold_for_case_attribute_values
			***** ignore_values_threshold_for_case_attributes
			***** ignore_values_threshold_for_event_attributes
	* '''columns''': An array of attribute column configuration objects used to define columns in the input data that are to be used as event- or case attributes.		* '''columns''': An array of attribute column configuration objects used to define columns in the input data that are to be used as event- or case attributes.
	** If null, all the columns will be included as categorical attributes (except case id, event type (only for event) and timestamp (only for event) columns).		** If null, all the columns will be included as categorical attributes (except case id, event type (only for event) and timestamp (only for event) columns).

Ollvihe: /* Create prediction script in QPR ProcessAnalyzer */

2025-12-02T12:41:36Z

Create prediction script in QPR ProcessAnalyzer

← Older revision		Revision as of 12:41, 2 December 2025
Line 37:		Line 37:
	1. Create the following example expression script (e.g., with name '''Create prediction model'''):<syntaxhighlight lang="typescript">		1. Create the following example expression script (e.g., with name '''Create prediction model'''):<syntaxhighlight lang="typescript">
	let sourceModel = ProjectByName("<project name>").ModelByName("<model name>");		let sourceModel = ProjectByName("<project name>").ModelByName("<model name>");
	let ~~completeCaseEventTypeName~~ = "<event type name found only in complete cases>";		let completeCaseEventTypeNames = ["<event type name found only in complete cases>", "<another event type name>", "..."];

	let targetProject = Project;		let targetProject = Project;
Line 60:		Line 60:
	"Type": "EventAttributeValue",		"Type": "EventAttributeValue",
	"Attribute": eventTypeColumnName,		"Attribute": eventTypeColumnName,
	"~~StringifiedValues~~": [		"Values": completeCaseEventTypeNames
	~~`0${completeCaseEventTypeName}`~~
	]
	}		}
	]		]
Line 76:		Line 74:
	"Type": "EventAttributeValue",		"Type": "EventAttributeValue",
	"Attribute": eventTypeColumnName,		"Attribute": eventTypeColumnName,
	"~~StringifiedValues~~": [		"Values": completeCaseEventTypeNames
	~~`0${completeCaseEventTypeName}`~~
	]
	}		}
	]		]

Ollvihe: /* Create prediction script in QPR ProcessAnalyzer */

2025-11-26T09:56:39Z

Create prediction script in QPR ProcessAnalyzer

← Older revision		Revision as of 09:56, 26 November 2025
Line 43:		Line 43:

	_system.ML.GeneratePredictionModel(#{		_system.ML.GeneratePredictionModel(#{
	"Name": "My prediction model", // Name of the PA model to generate ti the target project.		"Name": "My prediction model", // Name of the PA model to generate to the target project.
	"SourceModel": sourceModel, // Snowflake-based PA model used for training the prediction model.		"SourceModel": sourceModel, // Snowflake-based PA model used for training the prediction model.
	"TargetProject": targetProject, // Target project to create the model into.		"TargetProject": targetProject, // Target project to create the model into.

MarHink: /* Predicting case attribute values */

2025-10-21T06:59:26Z

Predicting case attribute values

← Older revision		Revision as of 06:59, 21 October 2025
Line 229:		Line 229:
	let caseAttributeToPredict = "<name of the case attribute>";		let caseAttributeToPredict = "<name of the case attribute>";
	let resultModelName = "<name of the model to be created/replaced>";		let resultModelName = "<name of the model to be created/replaced>";
	let generateDebugCaseAttributes = false;		let generateDebugCaseAttributes = false; // Set to true to generate columns for prediction probabilities.
	let casesToPredictFilter = "<JSON filter for cases for which the prediction is to be performed>";		let casesToPredictFilter = "<JSON filter for cases for which the prediction is to be performed>";
	let casesToUseForTrainingFilter = "<JSON filter for cases to be used for ML model training>";		let casesToUseForTrainingFilter = "<JSON filter for cases to be used for ML model training>";

MarHink: Added example of predicting case attribute values

2025-10-21T06:58:09Z

Added example of predicting case attribute values

@@ Line 222: / Line 222: @@
+   }
+}
 </syntaxhighlight>

MarHink: /* Install prediction to Snowflake */

2024-11-15T15:24:19Z

Install prediction to Snowflake

← Older revision		Revision as of 15:24, 15 November 2024
Line 12:		Line 12:
	== Install prediction to Snowflake ==		== Install prediction to Snowflake ==
	To install the eventlog prediction to Snowflake:		To install the eventlog prediction to Snowflake:
	# Go to Snowflake, and create a Snowflake-managed stage with name '''~~PREDICTION~~''' to the same schema configured to QPR ProcessAnalyzer (in the Snowflake connection string). Use settings in the following image: [[File:Create_Snowflake_stage.png]]		# Go to Snowflake, and create a Snowflake-managed stage with name '''DECISION_INTELLIGENCE''' to the same schema configured to QPR ProcessAnalyzer (in the Snowflake connection string). Use settings in the following image: [[File:Create_Snowflake_stage.png]]
	# Open the created stage and upload the '''predict.pyz''' file into the stage (ask the file from your QPR representative).		# Open the created stage and upload the '''predict.pyz''' file into the stage (ask the file from your QPR representative).
	# Create the following procedure to the same schema:		# Create the following procedure to the same schema:

MarHink: /* Install prediction to Snowflake */

2024-11-15T15:22:53Z

Install prediction to Snowflake

← Older revision		Revision as of 15:22, 15 November 2024
Line 16:		Line 16:
	# Create the following procedure to the same schema:		# Create the following procedure to the same schema:
	<syntaxhighlight lang="sql">		<syntaxhighlight lang="sql">
	CREATE OR REPLACE PROCEDURE ~~MARKKU_HINKKA.QPRPA.QPRPA_SP_PREDICTION_MARHINK~~("CONFIGURATION" OBJECT)		CREATE OR REPLACE PROCEDURE QPRPA_SP_PREDICTION("CONFIGURATION" OBJECT)
	RETURNS OBJECT		RETURNS OBJECT
	LANGUAGE PYTHON		LANGUAGE PYTHON
Line 27:		Line 27:
	import sys		import sys
	def main(session, parameters_in: dict) -> dict:		def main(session, parameters_in: dict) -> dict:
	session.file.get(''@decision_intelligence/~~predict_marhink~~.pyz'', ''/tmp'')		session.file.get(''@decision_intelligence/predict.pyz'', ''/tmp'')
	sys.path.append(''/tmp/~~predict_marhink~~.pyz'')		sys.path.append(''/tmp/predict.pyz'')
	import predict		import predict
	return predict.main(session, parameters_in)		return predict.main(session, parameters_in)

MarHink: /* Attribute configuration */

2024-11-15T14:48:12Z

Attribute configuration

← Older revision		Revision as of 14:48, 15 November 2024
Line 156:		Line 156:

	== Attribute configuration ==		== Attribute configuration ==
	Attribute configuration is used to configure which event- and case attributes should be used in prediction model and how they are used.		Attribute configuration is used in '''TrainingConfiguration''' (see the chapter above) to configure which event- and case attributes should be used in prediction model and how they are used.

	The configuration is in the top level split into two sections: "event" and "case". "Event" is used to configure event attributes, whereas "case" is used for case attributes.		The configuration is in the top level split into two sections: "event" and "case". "Event" is used to configure event attributes, whereas "case" is used for case attributes.