nb
Class NaiveBayesInd

java.lang.Object
  |
  +--shared.BaseInducer
        |
        +--shared.Inducer
              |
              +--nb.NaiveBayesInd

public class NaiveBayesInd

extends Inducer

Field Summary

java.lang.String	endl
java.lang.String	EVIDENCE_FACTOR_HELP
java.lang.String	KL_THREHSOLD_HELP
java.lang.String	LAPLACE_HELP
java.lang.String	M_FACTOR_HELP
java.lang.String	UNKNOWN_IS_VALUE_HELP
static int	unknownAuto
static int	unknownNo
static int	unknownYes
java.lang.String	USE_EVIDENCE_PROJECTION_HELP

Fields inherited from class shared.BaseInducer

AHA_IB_INDUCER, AM_INDUCER, BAGGING_INDUCER, BOOSTER_INDUCER, C45_INDUCER, C45AP_INDUCER, C45R_INDUCER, C50_INDUCER, CART_INDUCER, CatDT_INDUCER, CF_INDUCER, CLUSTER_INDUCER, CN2_INDUCER, CONST_INDUCER, COODG_INDUCER, DDT_INDUCER, DF_INDUCER, DISC_NB_INDUCER, DISC_SEARCH_INDUCER, DISC_TAB_INDUCER, ENTROPY_ODG_INDUCER, FCF_INDUCER, FSS_INDUCER, getEnv, HOODG_INDUCER, IB_INDUCER, ID3_INDUCER, LAZY_DT_INDUCER, LIST_HOODG_INDUCER, LIST_ODG_INDUCER, logOptions, NAIVE_BAYES_INDUCER, NULL_INDUCER, OC1_INDUCER, ODT_INDUCER, ONER_INDUCER, OODG_INDUCER, ORDER_FSS_INDUCER, PEBLS_INDUCER, PERCEPTRON_INDUCER, PERF_EST_INDUCER, PROJECT_INDUCER, RIPPER_INDUCER, SGI_DT_INDUCER, STACKING_INDUCER, T2_INDUCER, TABLE_CAS_INDUCER, TABLE_INDUCER, TDDT_INDUCER, TS, WEIGHT_SEARCH_INDUCER, WINNOW_INDUCER

Constructor Summary

NaiveBayesInd(NaiveBayesInd source)
Description : Copy constructor.

NaiveBayesInd(java.lang.String description)
Constructor with description String.

Method Summary

int	class_id() returns the Id of this inducer
Inducer	copy() Description : Returns the pointer to the copy of this.
void	display(java.io.BufferedWriter stream, DisplayPref dp) Description : Prints a readable representation of the Cat to the given stream.
Categorizer	get_categorizer() Description : Returns the categorizer that the inducer has generated.
double	get_m_estimate_factor()
boolean	get_use_evidence_projection()
boolean	get_use_laplace()
int	num_nontrivial_leaves() This method returns 0 because NaiveBayes is not a tree inducer.
int	num_nontrivial_nodes() This method returns 0 because NaiveBayes is not a tree inducer.
Categorizer	release_categorizer() Description : Gives ownership of the generated categorizer to the caller, reverting the Inducer to untrained state.
void	set_m_estimate_factor(double m) Description : set m value for L'aplace correction.
void	set_use_evidence_projection(boolean projection)
void	set_use_laplace(boolean laplace) Description : set whether to use L'aplace correction.
void	set_user_options(java.lang.String preFix) Description : Sets the options from environment variables.
void	train() Description : Trains Naive Bayes Categorizer.
boolean	was_trained(boolean fatal_on_false) Description : Return TRUE iff the class has a valid categorizer.

Methods inherited from class shared.Inducer

can_cast_to_inducer, cast_to_inducer, display_struct, display_struct, project_train_and_perf_files, project_train_and_perf_files, project_train_and_perf_files, project_train_and_perf_files, project_train_and_perf, project_train_and_test_files, project_train_and_test_files, project_train_and_test_files, project_train_and_test_files, project_train_and_test, supports_full_testing, train_and_perf, train_and_test

Methods inherited from class shared.BaseInducer

assign_data, can_cast_to_incr_inducer, cast_to_incr_inducer, description, get_log_level, get_log_options, get_log_stream, has_data, has_data, instance_list, normalize_weights, read_data, read_data, read_data, release_data, set_log_level, set_log_options, set_log_prefixes, set_log_stream, train_and_test_files, train_and_test_files, train_and_test_files, train_and_test_files

Methods inherited from class java.lang.Object

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

Field Detail

endl

public java.lang.String endl

LAPLACE_HELP

public final java.lang.String LAPLACE_HELP

M_FACTOR_HELP

public final java.lang.String M_FACTOR_HELP

UNKNOWN_IS_VALUE_HELP

public final java.lang.String UNKNOWN_IS_VALUE_HELP

KL_THREHSOLD_HELP

public final java.lang.String KL_THREHSOLD_HELP

USE_EVIDENCE_PROJECTION_HELP

public final java.lang.String USE_EVIDENCE_PROJECTION_HELP

EVIDENCE_FACTOR_HELP

public final java.lang.String EVIDENCE_FACTOR_HELP

unknownNo

public static final int unknownNo

unknownYes

public static final int unknownYes

unknownAuto

public static final int unknownAuto

Constructor Detail

NaiveBayesInd

public NaiveBayesInd(java.lang.String description)

Constructor with description String.

NaiveBayesInd

public NaiveBayesInd(NaiveBayesInd source)

Description : Copy constructor. Copy correction options. Comments :

Method Detail

class_id

public int class_id()

returns the Id of this inducer

Overrides:

class_id in class BaseInducer

Tags copied from class: BaseInducer

Returns:

Integer assigned to this inducer.

was_trained

public boolean was_trained(boolean fatal_on_false)

Description : Return TRUE iff the class has a valid categorizer. Comments :

get_categorizer

public Categorizer get_categorizer()

Description : Returns the categorizer that the inducer has generated. Comments :

Overrides:

get_categorizer in class Inducer

Tags copied from class: Inducer

Returns:

The Categorizer class storing a categorizer created by this induction algorithm.

release_categorizer

public Categorizer release_categorizer()

Description : Gives ownership of the generated categorizer to the caller, reverting the Inducer to untrained state. Comments :

Overrides:

release_categorizer in class Inducer

Tags copied from class: Inducer

Returns:

The Categorizer trained.

set_user_options

public void set_user_options(java.lang.String preFix)

Description : Sets the options from environment variables. Comments :

train

public void train()

Description : Trains Naive Bayes Categorizer. Descrete attributes can be handled by simply passing occurance counts in the BagCounter. Continuous attributes are fed into StatDatas to get mean, variance, and standard deviation. Comments : It is possible that some label is not in the training set, or it has unknown for a continous attribute. In this case, statData::size() = 0, and the loop below will not write any data into the NBNorm structure. Since the NBNorm structure is initialized to HasData = FALSE, doing nothing will correctly indicate that there is no data.

Overrides:

train in class Inducer

display

public void display(java.io.BufferedWriter stream,
                    DisplayPref dp)
             throws java.io.IOException

Description : Prints a readable representation of the Cat to the given stream. Comments :

copy

public Inducer copy()

Description : Returns the pointer to the copy of this. Comments :

set_use_laplace

public void set_use_laplace(boolean laplace)

Description : set whether to use L'aplace correction. Comments :

set_m_estimate_factor

public void set_m_estimate_factor(double m)

Description : set m value for L'aplace correction. Comments :

get_use_laplace

public boolean get_use_laplace()

get_m_estimate_factor

public double get_m_estimate_factor()

get_use_evidence_projection

public boolean get_use_evidence_projection()

set_use_evidence_projection

public void set_use_evidence_projection(boolean projection)

num_nontrivial_nodes

public int num_nontrivial_nodes()

This method returns 0 because NaiveBayes is not a tree inducer. There is no tree for which to count nodes.

Returns:

0;

num_nontrivial_leaves

public int num_nontrivial_leaves()

This method returns 0 because NaiveBayes is not a tree inducer. There is not tree for which to count leaves.

Returns:

0