org.knime.base.node.mine.bayes.naivebayes.datamodel
Class NumericalAttributeModel

java.lang.Object
  org.knime.base.node.mine.bayes.naivebayes.datamodel.AttributeModel
      org.knime.base.node.mine.bayes.naivebayes.datamodel.NumericalAttributeModel

All Implemented Interfaces:

Comparable<AttributeModel>

 class NumericalAttributeModel
extends AttributeModel

This AttributeModel implementation calculates the probability for numerical attributes by assuming a Gaussian distribution of the data.

Author:

Tobias Koetter, University of Konstanz

Field Summary

(package private) static String

MODEL_TYPE The unique type of this model used for saving/loading.

Fields inherited from class org.knime.base.node.mine.bayes.naivebayes.datamodel.AttributeModel

MODEL_CONTAINS_NO_CLASS_VALUES, MODEL_CONTAINS_NO_RECORDS

Constructor Summary

NumericalAttributeModel(String attributeName, boolean skipMissingVals)
Constructor for class NumericalRowValue.

NumericalAttributeModel(String attributeName, boolean skipMissingVals, int noOfMissingVals, Config config)
Constructor for class NumericalAttributeModel.

Method Summary

(package private) void	addValueInternal(String classValue, DataCell attrValue) Adds the given value to the concrete implementation.
(package private) Collection<String>	getClassValues()
(package private) Class<? extends DataValue>	getCompatibleType()
(package private) String	getHTMLView(int totalNoOfRecs)
(package private) String	getHTMLViewHeadLine()
(package private) Integer	getNoOfRecs4ClassValue(String classValue)
(package private) double	getProbabilityInternal(String classValue, DataCell attributeValue, double laplaceCorrector) This should also handle missing values.
(package private) String	getType()
(package private) void	saveModelInternal(Config config)
String	toString()
(package private) void	validate() Called after all training rows where added to validate the model.

Methods inherited from class org.knime.base.node.mine.bayes.naivebayes.datamodel.AttributeModel

addValue, compareTo, createHTMLTable, createTableHeader, getAttributeName, getInvalidCause, getMissingValueHeader, getNoOfMissingVals, getProbability, loadModel, saveModel, setInvalidCause, sortCollection

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, wait, wait, wait

Field Detail

MODEL_TYPE

static final String MODEL_TYPE

The unique type of this model used for saving/loading.

See Also:

Constant Field Values

Constructor Detail

NumericalAttributeModel

NumericalAttributeModel(String attributeName,
                        boolean skipMissingVals)

Constructor for class NumericalRowValue.

Parameters:

attributeName - the row caption

skipMissingVals - set to true if the missing values should be skipped during learning and prediction

NumericalAttributeModel

NumericalAttributeModel(String attributeName,
                        boolean skipMissingVals,
                        int noOfMissingVals,
                        Config config)
                  throws InvalidSettingsException

Constructor for class NumericalAttributeModel.

Parameters:

attributeName - the name of the attribute

skipMissingVals - set to true if the missing values should be skipped during learning and prediction

noOfMissingVals - the number of missing values

config - the Config object to read from

Throws:

InvalidSettingsException - if the settings are invalid

Method Detail

saveModelInternal

void saveModelInternal(Config config)

Specified by:

saveModelInternal in class AttributeModel

Parameters:

config - the config object to save to

addValueInternal

void addValueInternal(String classValue,
                      DataCell attrValue)

Adds the given value to the concrete implementation. Should handle missing values as well.

Specified by:

addValueInternal in class AttributeModel

Parameters:

classValue - the class value

attrValue - the attribute value. Could be a missing value.

validate

void validate()
        throws InvalidSettingsException

Called after all training rows where added to validate the model.

Specified by:

validate in class AttributeModel

Throws:

InvalidSettingsException - if the model isn't valid

getCompatibleType

Class<? extends DataValue> getCompatibleType()

Specified by:

getCompatibleType in class AttributeModel

Returns:

the DataValue class to check if the rows are compatible

getClassValues

Collection<String> getClassValues()

Specified by:

getClassValues in class AttributeModel

Returns:

all class values

getNoOfRecs4ClassValue

Integer getNoOfRecs4ClassValue(String classValue)

Specified by:

getNoOfRecs4ClassValue in class AttributeModel

Parameters:

classValue - the class value we want the number of records for

Returns:

the number of records with the given class value or null if only missing values where in this row for the given class value or the class value wasn't found at all

getProbabilityInternal

double getProbabilityInternal(String classValue,
                              DataCell attributeValue,
                              double laplaceCorrector)

This should also handle missing values.

Specified by:

getProbabilityInternal in class AttributeModel

Parameters:

classValue - the class value to calculate the probability for

attributeValue - the attribute value to calculate the probability for. Could be a missing value.

laplaceCorrector - the Laplace corrector to use. A value greater 0 overcomes zero counts.

Returns:

the calculated probability

getHTMLViewHeadLine

String getHTMLViewHeadLine()

Specified by:

getHTMLViewHeadLine in class AttributeModel

Returns:

the headline of this model to use in the HTML view

getType

String getType()

Specified by:

getType in class AttributeModel

Returns:

the unique type of the model

getHTMLView

String getHTMLView(int totalNoOfRecs)

Specified by:

getHTMLView in class AttributeModel

Parameters:

totalNoOfRecs - the total number of records in the training data

Returns:

the HTML view of this attribute model

toString

public String toString()

Overrides:

toString in class Object