Class RuleGeneratorParams
java.lang.Object
adaa.analytics.rules.logic.rulegenerator.RuleGeneratorParams
-
Nested Class Summary
Nested ClassesModifier and TypeClassDescriptionprotected static enum
Auxiliary enumeration type describing possible destinations of quality measures. -
Field Summary
FieldsModifier and TypeFieldDescriptionstatic final String
static final String
static final String
static final String
Boolean indicating whether automatic induction should be performed for classes for which no user's knowledge has been defined (classification only).static final String
static final String
Binary parameter indicating whether pruning should be enabled.static final String
static final String
Set of forbidden conditions (used also for specifying forbidden attributes by using special value Any).static final String
Multiset of preferred conditions (used also for specifying preferred attributes by using special value "Any").static final String
Set of initial (expert's) rules.static final String
Auxiliary parameter for specifying sets/multisets of expert rules and preferred/forbidden conditions/attributes.static final String
Auxiliary parameter for specifying sets/multisets of expert rules and preferred/forbidden conditions/attributes.static final String
Boolean indicating whether initial rules should be extended with a use of automatic conditions.static final String
Boolean indicating whether initial rules should be extended with a use of preferred conditions and attributes.static final String
Boolean telling whether missing values should be ignored (by default, a missing value of given attribute is always considered as not fulfilling the condition build upon that attribute)static final String
static final String
Boolean indicating whether new rules should be induced with a use of automatic conditions.static final String
Boolean indicating whether new rules should be induced with a use of preferred conditions and attributes.static final String
Name of the rule quality measure used during growing (ignored in the survival analysis where log-rank statistics is used).static final String
Non-negative integer representing maximum number of conditions which can be added to the rule in the growing phase (use this parameter for large datasets if execution time is prohibitive); 0 indicates no limit.static final String
static final String
Max number of rules to generate - overrides min_cov settingstatic final String
Fraction of examples that may remain uncovered by the rule set.static final String
static final String
static final String
Number/fraction examples to be covered by a new rule.static final String
Number/fraction of previously uncovered examples to be covered by a new rule.static final String
static final String
static final String
Maximum number of preferred attributes per rule.static final String
Maximum number of preferred conditions per rule.static final String
Name of the rule quality measure used during pruning.static final String
Flag determining if best candidate should be selected from growing phase."static final String
Boolean indicating whether user's knowledge should be used.static final String
Class name of user-defined induction measure; applies only when the corresponding measure parameter has value UserDefined; the equation must be a mathematical expression with p, n, P, N literals (elements of confusion matrix), operators, numbers, and library functions (sin, log, etc.).static final String
Class name of user-defined pruning measure.static final String
Class name of user-defined voting measure.static final String
Name of the rule quality measure used during growing.Instead of parameter in settings (user_induction_class) one can set object with IUserMeasureInstead of parameter in settings (user_purning_class) one can set object with IUserMeasureInstead of parameter in settings (user_voting_class) one can set object with IUserMeasure -
Constructor Summary
Constructors -
Method Summary
Modifier and TypeMethodDescriptionboolean
getParameter
(String key) boolean
double
int
getParameterAsInt
(String key) getParameterList
(String key) void
setListParameter
(String key, List<String[]> o) void
setParameter
(String key, String o) void
setUserMeasureInductionObject
(IUserMeasure userMeasureInductionObject) void
setUserMeasurePurningObject
(IUserMeasure userMeasurePurningObject) void
setUserMeasureVotingObject
(IUserMeasure userMeasureVotingObject)
-
Field Details
-
PARAMETER_USE_EXPERT
Boolean indicating whether user's knowledge should be used.- See Also:
-
PARAMETER_EXPERT_CONFIG
- See Also:
-
PARAMETER_EXPERT_RULES
Set of initial (expert's) rules.- See Also:
-
PARAMETER_EXPERT_PREFERRED_CONDITIONS
Multiset of preferred conditions (used also for specifying preferred attributes by using special value "Any").- See Also:
-
PARAMETER_EXPERT_FORBIDDEN_CONDITIONS
Set of forbidden conditions (used also for specifying forbidden attributes by using special value Any).- See Also:
-
PARAMETER_EXPORT_KEY
Auxiliary parameter for specifying sets/multisets of expert rules and preferred/forbidden conditions/attributes.- See Also:
-
PARAMETER_EXPORT_VALUE
Auxiliary parameter for specifying sets/multisets of expert rules and preferred/forbidden conditions/attributes.- See Also:
-
PARAMETER_EXTEND_USING_PREFERRED
Boolean indicating whether initial rules should be extended with a use of preferred conditions and attributes.- See Also:
-
PARAMETER_EXTEND_USING_AUTOMATIC
Boolean indicating whether initial rules should be extended with a use of automatic conditions.- See Also:
-
PARAMETER_INDUCE_USING_PREFERRED
Boolean indicating whether new rules should be induced with a use of preferred conditions and attributes.- See Also:
-
PARAMETER_INDUCE_USING_AUTOMATIC
Boolean indicating whether new rules should be induced with a use of automatic conditions.- See Also:
-
PARAMETER_CONSIDER_OTHER_CLASSES
Boolean indicating whether automatic induction should be performed for classes for which no user's knowledge has been defined (classification only).- See Also:
-
PARAMETER_PREFERRED_CONDITIONS_PER_RULE
Maximum number of preferred conditions per rule.- See Also:
-
PARAMETER_PREFERRED_ATTRIBUTES_PER_RULE
Maximum number of preferred attributes per rule.- See Also:
-
PARAMETER_MINCOV_NEW
Number/fraction of previously uncovered examples to be covered by a new rule. (positive examples for classification problems).- See Also:
-
PARAMETER_MAX_RULE_COUNT
Max number of rules to generate - overrides min_cov setting- See Also:
-
PARAMETER_MINCOV_ALL
Number/fraction examples to be covered by a new rule. (positive examples for classification problems).- See Also:
-
PARAMETER_MAX_UNCOVERED_FRACTION
Fraction of examples that may remain uncovered by the rule set.- See Also:
-
PARAMETER_MAX_GROWING
Non-negative integer representing maximum number of conditions which can be added to the rule in the growing phase (use this parameter for large datasets if execution time is prohibitive); 0 indicates no limit.- See Also:
-
PARAMETER_SELECT_BEST_CANDIDATE
Flag determining if best candidate should be selected from growing phase."- See Also:
-
PARAMETER_INDUCTION_MEASURE
Name of the rule quality measure used during growing (ignored in the survival analysis where log-rank statistics is used).- See Also:
-
PARAMETER_ENABLE_PRUNING
Binary parameter indicating whether pruning should be enabled.- See Also:
-
PARAMETER_PRUNING_MEASURE
Name of the rule quality measure used during pruning.- See Also:
-
PARAMETER_VOTING_MEASURE
Name of the rule quality measure used during growing.- See Also:
-
PARAMETER_USER_INDUCTION_CLASS
Class name of user-defined induction measure; applies only when the corresponding measure parameter has value UserDefined; the equation must be a mathematical expression with p, n, P, N literals (elements of confusion matrix), operators, numbers, and library functions (sin, log, etc.).- See Also:
-
PARAMETER_USER_PRUNING_CLASS
Class name of user-defined pruning measure.- See Also:
-
PARAMETER_USER_VOTING_CLASS
Class name of user-defined voting measure.- See Also:
-
PARAMETER_IGNORE_MISSING
Boolean telling whether missing values should be ignored (by default, a missing value of given attribute is always considered as not fulfilling the condition build upon that attribute)- See Also:
-
PARAMETER_MAXCOV_NEGATIVE
- See Also:
-
PARAMETER_PENALTY_STRENGTH
- See Also:
-
PARAMETER_PENALTY_SATURATION
- See Also:
-
PARAMETER_MAX_PASSES_COUNT
- See Also:
-
PARAMETER_INCLUDE_BINARY_CONTRAST
- See Also:
-
PARAMETER_COMPLEMENTARY_CONDITIONS
- See Also:
-
PARAMETER_MEAN_BASED_REGRESSION
- See Also:
-
PARAMETER_CONTROL_APRORI_PRECISION
- See Also:
-
PARAMETER_APPROXIMATE_INDUCTION
- See Also:
-
PARAMETER_APPROXIMATE_BINS_COUNT
- See Also:
-
userMeasureInductionObject
Instead of parameter in settings (user_induction_class) one can set object with IUserMeasure -
userMeasurePurningObject
Instead of parameter in settings (user_purning_class) one can set object with IUserMeasure -
userMeasureVotingObject
Instead of parameter in settings (user_voting_class) one can set object with IUserMeasure
-
-
Constructor Details
-
RuleGeneratorParams
public RuleGeneratorParams()
-
-
Method Details
-
contains
-
getParameters
-
setParameter
-
setListParameter
-
getParameterAsBoolean
-
setUserMeasureInductionObject
-
setUserMeasurePurningObject
-
setUserMeasureVotingObject
-
getParameterAsInt
-
getParameterAsDouble
-
getParameterAsString
-
getParameter
-
getParameterList
-
toJsonString
-