javax.datamining.supervised.classification
Interface ClassificationModel

All Superinterfaces:

MiningObject, Model, SupervisedModel

public interface ClassificationModel

extends SupervisedModel

A ClassificationModel contains the metadata resulting from a model build using ClassificationSettings.

Classification in data mining is used to predict the class of a case with unknown class. Classification has been used in customer segmentation, business modeling, and credit analysis. As a type of supervised learning, an algorithm supporting classification builds a model from a set of predictors that are used to predict a target. The input or build data for a supervised learning algorithm requires the presence of both predictors and target attribute in each case. Given a pre-determined set of classes in the target attribute, classification analyzes the build data to determine to which class a given case belongs.

Author:

JSR-73 Java Data Mining Expert Group

See Also:

MiningObject, Model, SupervisedModel

Method Summary

double	getClassificationError() Returns the percentage, expressed between 0 and 100, of the incorrect predictions made by the model, if ClassificationSettingsFactory.supportsCapability(ClassificationCapability.classificationError) returns true.
CategorySet	getTargetCategorySet() Returns the set of target categories used in the model.
boolean	wasCostMatrixUsed() Returns true if a cost matrix was specified and used to build the model.

Methods inherited from interface javax.datamining.supervised.SupervisedModel

getTargetAttributeName

Methods inherited from interface javax.datamining.base.Model

getApplicationName, getAttributeStatistics, getBuildDuration, getBuildSettings, getEffectiveBuildSettings, getMajorVersion, getMiningAlgorithm, getMiningFunction, getMinorVersion, getModelDetail, getProviderName, getProviderVersion, getSignature, getTaskIdentifier, getUniqueIdentifier, getVersion

Methods inherited from interface javax.datamining.MiningObject

getCreationDate, getCreatorInfo, getDescription, getName, getObjectIdentifier, getObjectType, setDescription

Method Detail

getClassificationError

public double getClassificationError()

Returns the percentage, expressed between 0 and 100, of the incorrect predictions made by the model, if ClassificationSettingsFactory.supportsCapability(ClassificationCapability.classificationError) returns true. Otherwise, the returned value is undefined.

Returns:

double

getTargetCategorySet

public CategorySet getTargetCategorySet()

Returns the set of target categories used in the model. This cannot be null.

This information is typically obtained from the category set specified with the LogicalAttribute. However, if this is not provided, the vendor may populate this explicitly from the data.

Returns:

CategorySet

wasCostMatrixUsed

public boolean wasCostMatrixUsed()

Returns true if a cost matrix was specified and used to build the model. Returns false otherwise.

Returns:

boolean