List of Components and Config Options

DL-Learner Components

Filter components by implemented interfaces:

Click on a component to get an overview on its configuration options.


Brute Force Learner

short name: bruteForce
version: 0.8
implements: LearningAlgorithm, ClassExpressionLearningAlgorithm

This component does not have configuration options.


CELOE

short name: celoe
version: 1.0
implements: LearningAlgorithm, ClassExpressionLearningAlgorithm
description: CELOE is an adapted and extended version of the OCEL algorithm applied for the ontology engineering use case. See http://jens-lehmann.org/files/2011/celoe.pdf for reference.

option name description type default value required?
maxExecutionTimeInSecondsAfterImprovement maximum execution of the algorithm in seconds int 0 false
startClass You can specify a start class for the algorithm. To do this, you have to use Manchester OWL syntax without using prefixes. Description owl:Thing false
noisePercentage the (approximated) percentage of noise within the examples double 0.0 false
terminateOnNoiseReached specifies whether to terminate when noise criterion is met boolean false false
writeSearchTree specifies whether to write a search tree boolean false false
filterDescriptionsFollowingFromKB If true, then the results will not contain suggestions, which already follow logically from the knowledge base. Be careful, since this requires a potentially expensive consistency check for candidate solutions. boolean false false
maxDepth maximum depth of description double 7 false
searchTreeFile file to use for the search tree String log/searchTree.txt false
maxExecutionTimeInSeconds maximum execution of the algorithm in seconds int 10 false
useMinimizer Specifies whether returned expressions should be minimised by removing those parts, which are not needed. (Basically the minimiser tries to find the shortest expression which is equivalent to the learned expression). Turning this feature off may improve performance. boolean true false
maxClassExpressionTestsAfterImprovement The maximum number of candidate hypothesis the algorithm is allowed after an improvement in accuracy (0 = no limit). The algorithm will stop afterwards. (The real number of tests can be slightly higher, because this criterion usually won’t be checked after each single test.) int 0 false
reuseExistingDescription If true, the algorithm tries to find a good starting point close to an existing definition/super class of the given class in the knowledge base. boolean false false
replaceSearchTree specifies whether to replace the search tree in the log file after each run or append the new search tree boolean false false
stopOnFirstDefinition algorithm will terminate immediately when a correct definition is found boolean false false
maxClassExpressionTests The maximum number of candidate hypothesis the algorithm is allowed to test (0 = no limit). The algorithm will stop afterwards. (The real number of tests can be slightly higher, because this criterion usually won’t be checked after each single test.) int 0 false
singleSuggestionMode Use this if you are interested in only one suggestion and your learning problem has many (more than 1000) examples. boolean false false
maxNrOfResults Sets the maximum number of results one is interested in. (Setting this to a lower value may increase performance as the learning algorithm has to store/evaluate/beautify less descriptions). int 10 false

ClassLearningProblem

short name: clp
version: 0.6
implements: LearningProblem

option name description type default value required?
accuracyMethod Specifies, which method/function to use for computing accuracy. Available measues are “pred_acc” (predictive accuracy), “fmeasure” (F measure), “generalised_fmeasure” (generalised F-Measure according to Fanizzi and d’Amato). String pred_acc false
classToDescribe class of which a description should be learned NamedClass true
checkConsistency whether to check for consistency of suggestions (when added to ontology) boolean true false
approxDelta The Approximate Delta double 0.05 false
betaSC beta index for F-measure in super class learning double 3.0 false
betaEq beta index for F-measure in definition learning double 1.0 false
useApproximations Use Approximations boolean false false

Disjunctive ELTL

short name: deltl
version: 0.5
implements: LearningAlgorithm, ClassExpressionLearningAlgorithm
description: Disjunctive ELTL is an algorithm based on the refinement operator in http://jens-lehmann.org/files/2009/el_ilp.pdf with support for disjunctions.

option name description type default value required?
tryFullCoverage If yes, then the algorithm tries to cover all positive examples. Note that while this improves accuracy on the testing set, it may lead to overfitting. boolean false false
treeSearchTimeSeconds Specifies how long the algorithm should search for a partial solution (a tree). double 1.0 false

ELTL

short name: eltl
version: 0.5
implements: LearningAlgorithm, ClassExpressionLearningAlgorithm
description: ELTL is an algorithm based on the refinement operator in http://jens-lehmann.org/files/2009/el_ilp.pdf.

option name description type default value required?
instanceBasedDisjoints Specifies whether to use real disjointness checks or instance based ones (no common instances) in the refinement operator. boolean true false

Fuzzy CELOE

short name: fceloe
version: 0.2
implements: LearningAlgorithm, ClassExpressionLearningAlgorithm
description: See Fuzzy DL-Learner paper published at ISDA 2011.

This component does not have configuration options.


Fuzzy OWL API Reasoner

short name: foar
version: 0.2
implements: ReasonerComponent

This component does not have configuration options.


FuzzyPosNegLPStandard

short name: fuzzyPosNeg
version: 0.2
implements: LearningProblem

This component does not have configuration options.


KB File

short name: kbfile
version: 0.8
implements: KnowledgeSource

option name description type default value required?
url URL pointer to the KB file String false
fileName relative or absolute path to KB file String false

OEHeuristicRuntime

short name: celoe_heuristic
version: 0.5
implements: OtherComponent

option name description type default value required?
startNodeBonus no description available double 0.1 false

OWL API Reasoner

short name: oar
version: 0.8
implements: ReasonerComponent

option name description type default value required?
reasonerType The name of the OWL APIReasoner to use {“fact”, “hermit”, “owllink”, “pellet”, “elk”, “cel”} String pellet false
owlLinkURL The URL to the owl server String false

OWL Class Expression Learner

short name: ocel
version: 1.2
implements: LearningAlgorithm, ClassExpressionLearningAlgorithm

This component does not have configuration options.


OWL File

short name: owlfile
version: 0.9
implements: KnowledgeSource

This component does not have configuration options.


PCELOE

short name: pceloe
version: 1.0
implements: LearningAlgorithm, ClassExpressionLearningAlgorithm
description: CELOE is an adapted and extended version of the OCEL algorithm applied for the ontology engineering use case. See http://jens-lehmann.org/files/2011/celoe.pdf for reference.

option name description type default value required?
maxExecutionTimeInSecondsAfterImprovement maximum execution of the algorithm in seconds int 0 false
startClass You can specify a start class for the algorithm. To do this, you have to use Manchester OWL syntax without using prefixes. Description owl:Thing false
noisePercentage the (approximated) percentage of noise within the examples double 0.0 false
nrOfThreads number of threads running in parallel int 2 false
terminateOnNoiseReached specifies whether to terminate when noise criterion is met boolean false false
writeSearchTree specifies whether to write a search tree boolean false false
filterDescriptionsFollowingFromKB If true, then the results will not contain suggestions, which already follow logically from the knowledge base. Be careful, since this requires a potentially expensive consistency check for candidate solutions. boolean false false
maxDepth maximum depth of description double 7 false
searchTreeFile file to use for the search tree String log/searchTree.txt false
maxExecutionTimeInSeconds maximum execution of the algorithm in seconds int 10 false
useMinimizer Specifies whether returned expressions should be minimised by removing those parts, which are not needed. (Basically the minimiser tries to find the shortest expression which is equivalent to the learned expression). Turning this feature off may improve performance. boolean true false
maxClassExpressionTestsAfterImprovement The maximum number of candidate hypothesis the algorithm is allowed after an improvement in accuracy (0 = no limit). The algorithm will stop afterwards. (The real number of tests can be slightly higher, because this criterion usually won’t be checked after each single test.) int 0 false
reuseExistingDescription If true, the algorithm tries to find a good starting point close to an existing definition/super class of the given class in the knowledge base. boolean false false
replaceSearchTree specifies whether to replace the search tree in the log file after each run or append the new search tree boolean false false
maxClassExpressionTests The maximum number of candidate hypothesis the algorithm is allowed to test (0 = no limit). The algorithm will stop afterwards. (The real number of tests can be slightly higher, because this criterion usually won’t be checked after each single test.) int 0 false
singleSuggestionMode Use this if you are interested in only one suggestion and your learning problem has many (more than 1000) examples. boolean false false
maxNrOfResults Sets the maximum number of results one is interested in. (Setting this to a lower value may increase performance as the learning algorithm has to store/evaluate/beautify less descriptions). int 10 false

PosNegLPStandard

short name: posNegStandard
version: 0.8
implements: LearningProblem

option name description type default value required?
approxDelta The Approximate Delta double 0.05 false
accuracyMethod Specifies, which method/function to use for computing accuracy. Available measues are “pred_acc” (predictive accuracy), “fmeasure” (F measure), “generalised_fmeasure” (generalised F-Measure according to Fanizzi and d’Amato). String predacc false
useApproximations Use Approximations boolean false false

PosNegLPStrict

short name: posNegStrict
version: 0.8
implements: LearningProblem

option name description type default value required?
errorPenalty penalty for errors (example can be inferred to belong to the negated concept class) double 3 false
accuracyPenalty penalty for incorrectness (example belongs neither to concept nor its negation) double 1 false

Random Guesser

short name: randomGuesser
version: 0.8
implements: LearningAlgorithm, ClassExpressionLearningAlgorithm

This component does not have configuration options.


SPARQL endpoint

short name: sparql
version: 0.2
implements: KnowledgeSource

option name description type default value required?
url no description available URL true
namedGraphs no description available List [] false
defaultGraphs no description available List [] false

SPARQL endpoint fragment

short name: sparqlfrag
version: 0.5
implements: KnowledgeSource

This component does not have configuration options.


data subPropertyOf axiom learner

short name: dplsubprop
version: 0.1
implements: LearningAlgorithm, AxiomLearningAlgorithm

option name description type default value required?
propertyToDescribe DatatypeProperty false

dataproperty domain axiom learner

short name: dpldomain
version: 0.1
implements: LearningAlgorithm, AxiomLearningAlgorithm

option name description type default value required?
propertyToDescribe DatatypeProperty false

dataproperty range learner

short name: dblrange
version: 0.1
implements: LearningAlgorithm, AxiomLearningAlgorithm

option name description type default value required?
propertyToDescribe DatatypeProperty false

disjoint classes learner

short name: cldisjoint
version: 0.1
implements: LearningAlgorithm, AxiomLearningAlgorithm, ClassExpressionLearningAlgorithm

option name description type default value required?
classToDescribe NamedClass false

disjoint dataproperty axiom learner

short name: dpldisjoint
version: 0.1
implements: LearningAlgorithm, AxiomLearningAlgorithm

option name description type default value required?
propertyToDescribe DatatypeProperty false

disjoint objectproperty axiom learner

short name: opldisjoint
version: 0.1
implements: LearningAlgorithm, AxiomLearningAlgorithm

option name description type default value required?
propertyToDescribe ObjectProperty false

efficient SPARQL fragment extractor

short name: sparqls
version: 0.1
implements: KnowledgeSource

option name description type default value required?
aboxfilter Filter for the tbox, can use variable ?s, ?p amd ?o String false
defaultGraphURI default graph URI String true
ontologySchemaUrls List of Ontology Schema URLs List true
sparqlQuery Sparql Query String false
tboxfilter Filter for the tbox, can use variable ?example and ?class String false
recursionDepth recursion depth int true
endpointURL URL of the SPARQL endpoint String true
instances List of the instances to use List true

equivalent dataproperty axiom learner

short name: dplequiv
version: 0.1
implements: LearningAlgorithm, AxiomLearningAlgorithm

option name description type default value required?
propertyToDescribe DatatypeProperty false

equivalent objectproperty axiom learner

short name: oplequiv
version: 0.1
implements: LearningAlgorithm, AxiomLearningAlgorithm

option name description type default value required?
propertyToDescribe ObjectProperty false

fast instance checker

short name: fic
version: 0.9
implements: ReasonerComponent

option name description type default value required?
forAllRetrievalSemantics This option controls how to interpret the all quantifier in forall r.C. The standard option isto return all those which do not have an r-filler not in C. The domain semantics is to use thosewhich are in the domain of r and do not have an r-filler not in C. The forallExists semantics is touse those which have at least one r-filler and do not have an r-filler not in C. String standard false
defaultNegation Whether to use default negation, i.e. an instance not being in a class means that it is in the negation of the class. boolean true false

functional dataproperty axiom learner

short name: dplfunc
version: 0.1
implements: LearningAlgorithm, AxiomLearningAlgorithm

option name description type default value required?
propertyToDescribe DatatypeProperty false

functional objectproperty axiom learner

short name: oplfunc
version: 0.1
implements: LearningAlgorithm, AxiomLearningAlgorithm

option name description type default value required?
propertyToDescribe ObjectProperty false

inversefunctional objectproperty axiom learner

short name: oplinvfunc
version: 0.1
implements: LearningAlgorithm, AxiomLearningAlgorithm

option name description type default value required?
propertyToDescribe ObjectProperty false

multiple criteria heuristic

short name: multiheuristic
version: 0.7
implements: OtherComponent

option name description type default value required?
expansionPenaltyFactor no description available double 0.02 false
negativeWeight no description available double 1.0 false
startNodeBonus no description available double 0.1 false
negationPenalty no description available int 0 false
gainBonusFactor no description available double 0.5 false
nodeChildPenalty no description available double 0.0001 false

object subPropertyOf axiom learner

short name: oplsubprop
version: 0.1
implements: LearningAlgorithm, AxiomLearningAlgorithm

option name description type default value required?
propertyToDescribe ObjectProperty false

objectproperty domain axiom learner

short name: opldomain
version: 0.1
implements: LearningAlgorithm, AxiomLearningAlgorithm

option name description type default value required?
propertyToDescribe ObjectProperty false

objectproperty range learner

short name: oplrange
version: 0.1
implements: LearningAlgorithm, AxiomLearningAlgorithm

option name description type default value required?
propertyToDescribe ObjectProperty false

positive only learning problem

short name: posonlylp
version: 0.6
implements: LearningProblem

This component does not have configuration options.


rho refinement operator

short name: rho
version: 0.8
implements: RefinementOperator

option name description type default value required?
instanceBasedDisjoints no description available boolean true false
useStringDatatypes no description available boolean false false
useNegation no description available boolean true false
useAllConstructor no description available boolean true false
disjointChecks no description available boolean true false
applyExistsFilter no description available boolean true false
useHasValueConstructor no description available boolean false false
useBooleanDatatypes no description available boolean true false
useExistsConstructor no description available boolean true false
dropDisjuncts no description available boolean false false
applyAllFilter no description available boolean true false
useDoubleDatatypes no description available boolean true false
useCardinalityRestrictions no description available boolean true false

simple subclass learner

short name: clsub
version: 0.1
implements: LearningAlgorithm, AxiomLearningAlgorithm, ClassExpressionLearningAlgorithm

option name description type default value required?
classToDescribe NamedClass true

symmetric objectproperty axiom learner

short name: oplsymm
version: 0.1
implements: LearningAlgorithm, AxiomLearningAlgorithm

option name description type default value required?
propertyToDescribe ObjectProperty false

synchronized rho refinement operator

short name: syncrho
version: 0.8
implements: RefinementOperator

option name description type default value required?
instanceBasedDisjoints no description available boolean true false
useStringDatatypes no description available boolean false false
useNegation no description available boolean true false
useAllConstructor no description available boolean true false
disjointChecks no description available boolean true false
applyExistsFilter no description available boolean true false
useHasValueConstructor no description available boolean false false
useBooleanDatatypes no description available boolean true false
useExistsConstructor no description available boolean true false
dropDisjuncts no description available boolean false false
applyAllFilter no description available boolean true false
useDoubleDatatypes no description available boolean true false
useCardinalityRestrictions no description available boolean true false

transitive objectproperty axiom learner

short name: opltrans
version: 0.1
implements: LearningAlgorithm, AxiomLearningAlgorithm

option name description type default value required?
propertyToDescribe ObjectProperty false