Abstract. This paper proposes the FOGA (Fuzzy Ontology Generation frAmework) for automatic generation of fuzzy ontology on uncertainty information. The FOGA framework comprises the following components:

FOGA:AFuzzyOntologyGeneration

FrameworkforScholarlySemanticWebThanhThoQuan1,SiuCheungHui1andTruHoangCao2

SchoolofComputerEngineering,NanyangTechnologicalUniversity,

Singapore

{PA0218164B,asschui}@ntu.edu.sg

2FacultyofInformationTechnology,HochiminhCityUniversityofTechnology,

HochiminhCity,Vietnam

tru@dit.hcmut.edu.vn1

Abstract.ThispaperproposestheFOGA(FuzzyOntologyGenerationfrAmework)forautomaticgenerationoffuzzyontologyonuncertaintyinformation.TheFOGAframeworkcomprisesthefollowingcomponents:FuzzyFormalConceptAnalysis,FuzzyConceptualClusteringandFuzzyOntologyGeneration.First,FuzzyFormalConceptAnalysisincorporatesfuzzylogicintoFormalConceptAnalysis(FCA)toformafuzzyconceptlattice.FuzzyConceptualClusteringthenconstructstheconcepthierar-chyfromthefuzzyconceptlattice.Finally,FuzzyOntologyGenerationgeneratesthefuzzyontologyfromtheconcepthierarchy.Inthispaper,wewillalsodiscusstheapplicationoftheFOGAframeworktogener-ateascholarlyontologyfortheScholarlySemanticWebfromacitationdatabase.TheperformanceoftheproposedFOGAframeworkisgivenbasedonthescholarlyontologygenerated.

Keywords:FormalConceptAnalysis,FuzzyLogic,ConceptualClustering,Ontol-ogyGeneration

1Introduction

Ontologyisaconceptualizationofadomainintoahumanunderstandable,butmachine-readableformatconsistingofentities,attributes,relationshipsandax-ioms[1].Ontologyusesclassestorepresentconcepts.Ontologyalsosupportstaxonomyandnon-taxonomyrelationsbetweenclasses.However,theconcep-tualformalismsupportedbytypicalontologymaynotbesu cienttorepresentuncertaintyinformationthatiscommonlyfoundinmanyapplicationdomains.Forexample,keywordsextractedfromscienti cpublicationscanbeusedtoin-ferthecorrespondingresearchareas,however,itisinappropriatetotreatallkeywordsequallyassomekeywordsmaybemoresigni cantthanothers.Inad-dition,itissometimesdi culttojudgewhetheradocumentbelongscompletelytoaresearchareaornot.

Totacklethistypeofproblems,onepossiblesolutionistoincorporatefuzzylogicintoontologytohandleuncertaintydata.Traditionally,fuzzyontologyis