oracle.dmt.jdm.featureextraction
Class OraFeatureExtractionApplySettings

java.lang.Object
oracle.dmt.jdm.OraDMObject
oracle.dmt.jdm.OraMiningObject
oracle.dmt.jdm.task.apply.OraApplySettings
oracle.dmt.jdm.featureextraction.OraFeatureExtractionApplySettings

All Implemented Interfaces:

javax.datamining.task.apply.ApplySettings, javax.datamining.MiningObject, oracle.dmt.jdm.OraPLSQLConstants

public class OraFeatureExtractionApplySettings

extends oracle.dmt.jdm.task.apply.OraApplySettings

A OraFeatureExtractionApplySettings captures a specification that prescribes the output of an apply task specific to a feature extraction model.

Field Summary

Fields inherited from class oracle.dmt.jdm.task.apply.OraApplySettings

ALL_PREDICTIONS_APPLY_OUTPUT, DEFAULT_APPLY_OUTPUT, MULTIPLE_ROW_ALL, MULTIPLE_ROW_DEFAULT, MULTIPLE_ROW_RANKED, SINGLE_ROW_AS_SPECIFIED, TARGET_OR_CLUID_APPLY_OUTPUT, TOP_PREDICTION_APPLY_OUTPUT

Fields inherited from class oracle.dmt.jdm.OraMiningObject

DESCRIPTION_DELIMITER

Fields inherited from interface oracle.dmt.jdm.OraPLSQLConstants

abns_max_build_minutes, abns_max_nb_predictors, abns_max_predictors, abns_model_type, abns_multi_feature, abns_naive_bayes, abns_single_feature, algo_adaptive_bayes_network, algo_ai_mdl, algo_ai_mdl2, algo_apriori_association_rules, algo_decision_tree, algo_kmeans, algo_naive_bayes, algo_name, algo_nonnegative_matrix_factor, algo_ocluster, algo_predictor_variance, algo_support_vector_machines, apply_cost_content, apply_nodeid_content, apply_pred_value_content, apply_probability_content, asso_max_rule_length, asso_min_confidence, asso_min_support, association, association_in_model, attribute_importance, clas_cost_table_name, clas_priors_table_name, classification, clus_num_clusters, clustering, feat_num_features, feature_extraction, kmns_block_growth, kmns_conv_tolerance, kmns_cosine, kmns_distance, kmns_euclidean, kmns_fast_cosine, kmns_iterations, kmns_min_pct_attr_support, kmns_num_bins, kmns_size, kmns_split_criterion, kmns_variance, nabs_pairwise_threshold, nabs_singleton_threshold, nmfs_conv_tolerance, nmfs_num_iterations, nmfs_random_seed, ocluster_max_buffer, ocluster_sensitivity, operator_equal, operator_equal_v, operator_greater_or_equal, operator_greater_or_equal_v, operator_greater_than, operator_greater_than_v, operator_in, operator_in_v, operator_less_or_equal, operator_less_or_equal_v, operator_less_than, operator_less_than_v, operator_not_equal, operator_not_equal_v, operator_not_in, operator_not_in_v, oracle_char_type, oracle_dm_nested_categoricals, oracle_dm_nested_numericals, oracle_float_type, oracle_number_type, oracle_varchar2_type, regression, svms_active_learning, svms_al_disable, svms_al_enable, svms_complexity_factor, svms_conv_tolerance, svms_epsilon, svms_gaussian, svms_kernel_cache_size, svms_kernel_function, svms_linear, svms_outlier_rate, svms_std_dev, tree_impurity_entropy, tree_impurity_gini, tree_impurity_metric, tree_impurity_metric_default, tree_term_max_depth, tree_term_max_depth_default, tree_term_max_depth_max, tree_term_max_depth_min, tree_term_max_surrogates_max, tree_term_max_surrogates_min, tree_term_minpct_node, tree_term_minpct_node_default, tree_term_minpct_node_max, tree_term_minpct_split, tree_term_minpct_split_default, tree_term_minpct_split_max, tree_term_minrec_node, tree_term_minrec_node_default, tree_term_minrec_split, tree_term_minrec_split_default

Method Summary

OraFeatureExtractionApplyContent	getContent(java.lang.String destinationAttrName) Returns the apply content type designated to the specified physical attribute.
OraFeatureExtractionApplyContent[]	getContents(int rank) Returns an array of apply content types to appear in the apply output for the specified rank.
OraFeatureExtractionApplyContent[]	getContentsByFeature(int featureIdentifier) Returns the types of apply content to be included in the apply output for the specified cluser ID.
OraFeatureExtractionApplyContent[]	getMappedContents() Returns an array of apply contents that are used for mapping.
java.lang.String	getMappedDestinationAttrName(java.lang.Integer featureIdentifier, OraFeatureExtractionApplyContent contentType) Returns the name of the physical attribute mapped to the specified feature identifier.
java.lang.String	getMappedDestinationAttrName(int featureRank, OraFeatureExtractionApplyContent contentType) Returns the name of the physical attribute mapped to the specified feature identifier.
int[]	getMappedFeatureIdentifiers() Returns an array of feature identifiers mapped by mapByFeatureIdentifier method.
java.lang.Integer	getRank(java.lang.String destinationAttrName) Returns the rank designated to the specified physical attribute.
java.lang.Integer	getRanks() Returns the number of predictions specified by mapByRank method.
boolean	isFromTop() Returns true if predictions are ordered from top, false if from bottom.
void	mapByFeature(OraFeatureExtractionApplyContent content, int featureId, java.lang.String destinationAttrName) Maps the specified feature identifier to a destination attribute regardless of its rank.
void	mapByRank(OraFeatureExtractionApplyContent content, java.lang.String[] destPhysAttrNameArray, boolean fromTop) Maps the specified generated values for the predictions within the specified ranks to a set of destination attributes.
void	mapFeatures(OraFeatureExtractionApplyContent content, java.lang.String baseDestPhysAttrName) Maps all predictions in the model and the specified content value to a set of named destination attributes.
void	mapTopFeature(OraFeatureExtractionApplyContent content, java.lang.String destPhysAttrName) Maps the prediction with the highest probability to appear in the apply output under the specified attribute name.

Methods inherited from class oracle.dmt.jdm.task.apply.OraApplySettings

getAttributeNames, getContentAttrValue, getFunction, getMappedContentsEnums, getObjectType, getSourceDestinationMap, isInputObject, resetMapping, setSourceDestinationMap, verify

Methods inherited from class oracle.dmt.jdm.OraMiningObject

doBeforeStore, getCreationDate, getCreatorInfo, getDescription, getName, getObjectIdentifier, saveObjectInDatabase, setDescription

Methods inherited from class oracle.dmt.jdm.OraDMObject

createException, createException, createRuntimeException, createRuntimeException, getLocalizedMessage, isConnectionOpen, logInfo, logSevere, logTrace, logTrace, unsupported, unsupported

Methods inherited from class java.lang.Object

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

Methods inherited from interface javax.datamining.MiningObject

getCreationDate, getCreatorInfo, getDescription, getName, getObjectIdentifier, setDescription

Method Detail

getContentsByFeature

public OraFeatureExtractionApplyContent[] getContentsByFeature(int featureIdentifier)
                                                        throws javax.datamining.JDMException

Returns the types of apply content to be included in the apply output for the specified cluser ID.

Parameters:

featureIdentifier -

Returns:

OraFeatureExtractionApplyContent[]

Throws:

javax.datamining.JDMException

getMappedContents

public OraFeatureExtractionApplyContent[] getMappedContents()

Returns an array of apply contents that are used for mapping.

Returns:

OraFeatureExtractionApplyContent[]

mapByRank

public void mapByRank(OraFeatureExtractionApplyContent content,
                      java.lang.String[] destPhysAttrNameArray,
                      boolean fromTop)
               throws javax.datamining.JDMException

Maps the specified generated values for the predictions within the specified ranks to a set of destination attributes. The number of ranks depends on the cardinality of destination attributes. For example, a user may want to map the predicted target values of the top 3 prediction (rank=1,2,3) to 3 destination attributes, respectively.

Parameters:

content - A system-generated content to be included in the apply output.

destPhysAttrNameArray - An array of destination attributes.

fromTop - Indicates top predictions with the highest probabilities if true, or bottom predictions with lowest probabilities if false. If true/false, the top/bottom prediction is assigned to the first destination attribute (with index 0).

Returns:

void

Throws:

javax.datamining.JDMException

mapByFeature

public void mapByFeature(OraFeatureExtractionApplyContent content,
                         int featureId,
                         java.lang.String destinationAttrName)
                  throws javax.datamining.JDMException

Maps the specified feature identifier to a destination attribute regardless of its rank. This method is used when a specific feature is of special interest.

Parameters:

content - A system-generated content to be included in the apply output.

featureId - The feature identifier to be mapped.

destinationAttrName -

Returns:

void

Throws:

javax.datamining.JDMException

mapTopFeature

public void mapTopFeature(OraFeatureExtractionApplyContent content,
                          java.lang.String destPhysAttrName)
                   throws javax.datamining.JDMException

Maps the prediction with the highest probability to appear in the apply output under the specified attribute name.

Throws an exception if the destination attribute is duplicate.

Parameters:

content - The content type of the top prediction.

destPhysAttrName - The attribute name of the content type for the top prediction.

Returns:

void

Throws:

javax.datamining.JDMException

mapFeatures

public void mapFeatures(OraFeatureExtractionApplyContent content,
                        java.lang.String baseDestPhysAttrName)
                 throws javax.datamining.JDMException

Maps all predictions in the model and the specified content value to a set of named destination attributes. The names are created in such a way that the base destination name is suffixed with a number that starts from 1.

If the same content is used, the previous mapping is replaced with the new one.

Parameters:

content -

baseDestPhysAttrName -

Returns:

void

Throws:

javax.datamining.JDMException

getRank

public java.lang.Integer getRank(java.lang.String destinationAttrName)
                          throws javax.datamining.JDMException

Returns the rank designated to the specified physical attribute. distinationAttrName must not be null.

Parameters:

destinationAttrName - The name of the physical attribute whose designated rank is to be returned.

Returns:

Integer

Throws:

javax.datamining.JDMException

getRanks

public java.lang.Integer getRanks()

Returns the number of predictions specified by mapByRank method. Returns null if no such method has been invoked.

Returns:

Integer

isFromTop

public boolean isFromTop()

Returns true if predictions are ordered from top, false if from bottom. Returns null</n> if mapByRank method has not been invoked.

Returns:

boolean

getMappedFeatureIdentifiers

public int[] getMappedFeatureIdentifiers()

Returns an array of feature identifiers mapped by mapByFeatureIdentifier method.

Returns:

int[]

getMappedDestinationAttrName

public java.lang.String getMappedDestinationAttrName(java.lang.Integer featureIdentifier,
                                                     OraFeatureExtractionApplyContent contentType)
                                              throws javax.datamining.JDMException

Returns the name of the physical attribute mapped to the specified feature identifier. CategoryValue must not be NULL. ContentType must not be NULL.

Parameters:

featureIdentifier - The target value to be represented as the destination attribute.

contentType - The content type to be represented as the destination attribute.

Returns:

String

Throws:

javax.datamining.JDMException

getMappedDestinationAttrName

public java.lang.String getMappedDestinationAttrName(int featureRank,
                                                     OraFeatureExtractionApplyContent contentType)
                                              throws javax.datamining.JDMException

Returns the name of the physical attribute mapped to the specified feature identifier.

Parameters:

featureRank - The rank of the feature for the given content type

contentType - The content type to be represented as the destination attribute.

Returns:

String

Throws:

javax.datamining.JDMException

getContent

public OraFeatureExtractionApplyContent getContent(java.lang.String destinationAttrName)
                                            throws javax.datamining.JDMException

Returns the apply content type designated to the specified physical attribute. Name must not be NULL.

Parameters:

destinationAttrName - The name of the physical attribute whose designated content type is to be returned.

Returns:

FeatureExtractionApplyContent

Throws:

javax.datamining.JDMException

getContents

public OraFeatureExtractionApplyContent[] getContents(int rank)
                                               throws javax.datamining.JDMException

Returns an array of apply content types to appear in the apply output for the specified rank. Rank must be >= 1.

Parameters:

rank - The rank of the prediction whose content types are to be returned.

Returns:

FeatureExtractionApplyContent[]

Throws:

javax.datamining.JDMException