weka.classifiers.lazy
Class IB1_Anon

java.lang.Object
  weka.classifiers.Classifier
      weka.classifiers.lazy.IB1_Anon

All Implemented Interfaces:

java.io.Serializable, java.lang.Cloneable, weka.classifiers.UpdateableClassifier, weka.core.CapabilitiesHandler, weka.core.OptionHandler, weka.core.TechnicalInformationHandler

public class IB1_Anon
extends weka.classifiers.Classifier
implements weka.classifiers.UpdateableClassifier, weka.core.TechnicalInformationHandler

Instance-based classification method for anonymized data as described in the following paper:

 @inproceedings{DBLP:conf/icde/InanKB09,
 author    = {Ali Inan and
             Murat Kantarcioglu and
             Elisa Bertino},
 title     = {Using Anonymized Data for Classification},
 booktitle = {ICDE},
 year      = {2009},
 pages     = {429-440}
 }
 

All quasi-identifier information is read from the configuration file that was used in anonymizing the original dataset. Please make sure that any parameters passed as program arguments during anonymization are reflected in the configuration file (especially the outputFormat field).

We assume that the configuration file did not contain any specifications for identifier attributes. Otherwise, the configuration file should be edited such that all reference to id attributes are removed and the indices of quasi-identifier attributes are re-set according to the new schema (e.g., if the index of a QID attribute was 4 and an id attribute was at index 2; the updated index of the QID attribute should become 3 to reflect the removal of the id attribute).

Generalized attributes are represented as attributes of type String within the WEKA framework. Such representation allows handling of complex generalizations. For example if QI-Statistics are provided instead of generalizations we can compare the two distributions. On the other hand, if generalized values are provided, we can set the mid-point for numeric attributes and treat generalizations as new categories. Another major advantage is that, the user does not have to deal with the tedious work of listing all possible generalizations in the ARFF file header (i.e., simply input "@ATTRIBUTE name string" for a quasi-identifier attribute with the name "name").

When comparing two generalized values, a distinct comparison function gen_distance is called. This function assumes that both generalized values are obtained from the same output format of the anonymization toolbox. If this assumption fails, gen_distance will print an error message and exit with an error code. Therefore it is important that the input records are all output in a similar fashion.

Among different output formats supported by the anonymization toolbox, this classifier can only handle genVals and genValsDist. For the time being, we do not plan add support for the anatomy format due to the costly join operation involved.

See Also:

Serialized Form

Field Summary

private java.lang.String	configFile path to the configuration file
private double[]	m_MaxArray The maximum values for numeric attributes.
private double[]	m_MinArray The minimum values for numeric attributes.
private weka.core.Instances	m_Train The training instances used for classification.
private QIDAttribute[]	qids QuasiIdentifiers for generalized attributes.
(package private) static long	serialVersionUID for serialization

Fields inherited from class weka.classifiers.Classifier

m_Debug

Constructor Summary

IB1_Anon()
Class constructor.

IB1_Anon(Configuration conf)
Class constructor.

Method Summary

void	buildClassifier(weka.core.Instances instances) Generates the classifier.
double	classifyInstance(weka.core.Instance instance) Classifies the given test instance.
java.lang.String	configFileTipText() Returns the tip text for this property
private double	distance(weka.core.Instance first, weka.core.Instance second) Calculates the distance between two instances
private double	gen_distance(java.lang.String val1, java.lang.String val2, QIDAttribute qid) Method of calculating the generalized distance between any two values.
weka.core.Capabilities	getCapabilities() Returns default capabilities of the classifier.
java.lang.String	getConfigFile() Get the anonymization configuration file path
java.lang.String[]	getOptions() Gets the current settings of IB1_Anon
weka.core.TechnicalInformation	getTechnicalInformation()
java.lang.String	globalInfo() Returns a string describing classifier
java.util.Enumeration	listOptions() Returns an enumeration describing the available options.
static void	main(java.lang.String[] argv) Main method for testing this class.
private double	norm(double x, int i) Normalizes a given value of a numeric attribute.
private void	setConfigFile(java.lang.String config)
void	setOptions(java.lang.String[] options) -config <num> path to the anonymization configuration file (default config.xml).
java.lang.String	toString() Returns a description of this classifier.
void	updateClassifier(weka.core.Instance instance) Updates the classifier.
private void	updateMinMax(weka.core.Instance instance) Updates the minimum and maximum values for all the attributes based on a new instance.

Methods inherited from class weka.classifiers.Classifier

debugTipText, distributionForInstance, forName, getDebug, makeCopies, makeCopy, runClassifier, setDebug

Methods inherited from class java.lang.Object

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

Field Detail

configFile

private java.lang.String configFile

path to the configuration file

serialVersionUID

static final long serialVersionUID

for serialization

See Also:

Constant Field Values

m_Train

private weka.core.Instances m_Train

The training instances used for classification.

m_MinArray

private double[] m_MinArray

The minimum values for numeric attributes.

m_MaxArray

private double[] m_MaxArray

The maximum values for numeric attributes.

qids

private QIDAttribute[] qids

QuasiIdentifiers for generalized attributes.

Constructor Detail

IB1_Anon

public IB1_Anon()

Class constructor.

Parameters:

confFilename - name of the configuration file, used for anonymization

IB1_Anon

public IB1_Anon(Configuration conf)

Class constructor.

Parameters:

conf - Anonymization configuration

Method Detail

globalInfo

public java.lang.String globalInfo()

Returns a string describing classifier

Returns:

a description suitable for displaying in the explorer/experimenter gui

listOptions

public java.util.Enumeration listOptions()

Returns an enumeration describing the available options.

Specified by:

listOptions in interface weka.core.OptionHandler

Overrides:

listOptions in class weka.classifiers.Classifier

Returns:

an enumeration of all the available options.

configFileTipText

public java.lang.String configFileTipText()

Returns the tip text for this property

Returns:

tip text for this propert

getConfigFile

public java.lang.String getConfigFile()

Get the anonymization configuration file path

Returns:

path to the configuration file

getOptions

public java.lang.String[] getOptions()

Gets the current settings of IB1_Anon

Specified by:

getOptions in interface weka.core.OptionHandler

Overrides:

getOptions in class weka.classifiers.Classifier

Returns:

an array of strings suitable for passing to setOptions()

setOptions

public void setOptions(java.lang.String[] options)
                throws java.lang.Exception

 -config <num>
 path to the anonymization configuration file
 (default config.xml).
 

Specified by:

setOptions in interface weka.core.OptionHandler

Overrides:

setOptions in class weka.classifiers.Classifier

Parameters:

options - the list of options as an array of strings

Throws:

java.lang.Exception

setConfigFile

private void setConfigFile(java.lang.String config)
                    throws java.lang.Exception

Throws:

java.lang.Exception

getTechnicalInformation

public weka.core.TechnicalInformation getTechnicalInformation()

Specified by:

getTechnicalInformation in interface weka.core.TechnicalInformationHandler

getCapabilities

public weka.core.Capabilities getCapabilities()

Returns default capabilities of the classifier.

Specified by:

getCapabilities in interface weka.core.CapabilitiesHandler

Overrides:

getCapabilities in class weka.classifiers.Classifier

Returns:

the capabilities of this classifier

buildClassifier

public void buildClassifier(weka.core.Instances instances)
                     throws java.lang.Exception

Generates the classifier.

Specified by:

buildClassifier in class weka.classifiers.Classifier

Parameters:

instances - set of instances serving as training data

Throws:

java.lang.Exception - if the classifier has not been generated successfully

updateClassifier

public void updateClassifier(weka.core.Instance instance)
                      throws java.lang.Exception

Updates the classifier.

Specified by:

updateClassifier in interface weka.classifiers.UpdateableClassifier

Parameters:

instance - the instance to be put into the classifier

Throws:

java.lang.Exception - if the instance could not be included successfully

classifyInstance

public double classifyInstance(weka.core.Instance instance)
                        throws java.lang.Exception

Classifies the given test instance.

Overrides:

classifyInstance in class weka.classifiers.Classifier

Parameters:

instance - the instance to be classified

Returns:

the predicted class for the instance

Throws:

java.lang.Exception - if the instance can't be classified

toString

public java.lang.String toString()

Returns a description of this classifier.

Overrides:

toString in class java.lang.Object

Returns:

a description of this classifier as a string.

distance

private double distance(weka.core.Instance first,
                        weka.core.Instance second)
                 throws java.lang.Exception

Calculates the distance between two instances

Parameters:

first - the first instance

second - the second instance

Returns:

the distance between the two given instances

Throws:

java.lang.Exception

gen_distance

private double gen_distance(java.lang.String val1,
                            java.lang.String val2,
                            QIDAttribute qid)
                     throws java.lang.Exception

Method of calculating the generalized distance between any two values.

Parameters:

val1 - String representation of the first generalized value

val2 - String representation of the second generalized value

qid - Quasi-identifier attribute

Returns:

Expected distance between two generalized values

Throws:

java.lang.Exception

norm

private double norm(double x,
                    int i)

Normalizes a given value of a numeric attribute.

Parameters:

x - the value to be normalized

i - the attribute's index

Returns:

the normalized value

updateMinMax

private void updateMinMax(weka.core.Instance instance)

Updates the minimum and maximum values for all the attributes based on a new instance.

Parameters:

instance - the new instance

main

public static void main(java.lang.String[] argv)

Main method for testing this class.

Parameters:

argv - should contain command line arguments for evaluation (see Evaluation).