IDmBackPropagation

Assembly: Ms;

Namespace: Prognoz.Platform.Interop.Ms;

Description

The IDmBackPropagation interface is used for missing data substitution using the back-propagation network.

Inheritance Hierarchy

IDmMethodDetails

IDmAttributesDetails

IDmTargetDetails

IDmTemplateFillDetails

IDmBackPropagation

Comments

The method uses neural back-propagation network to substitute missing data. This network uses a learning algorithm where the error propagates through the network from the output layer to the input one: that is, in the direction opposite to the direction of signal propagation when the network functions normally. The method works on the numeric values.

Properties inherited from IDmTargetDetails

	Property name	Brief description
	Target	The Target property determines the field index of the data source, containing analyzed information.

Properties inherited from IDmAttributesDetails

	Property name	Brief description
	Attributes	The Attributes property determines the field indexes of the data source, containing factors for the analysis.

Properties inherited from IDmMethodDetails

	Property name	Brief description
	CrossValidation	The CrossValidation property returns cross-validation settings
	DisplayName	The DisplayName property returns the method name.
	StatMethod	The StatMethod property returns statistical method parameters used in analysis.

Methods inherited from IDmTemplateFillDetails

	Method name	Brief description
	FillTarget	The FillTarget method unloads missing data treatment results into the data source