Dataset about om2008-alignments.
Tue May 03 19:04:15 CEST 2016
DSSim results for OAEI 2008
DSSim results for OAEI 2008
DSSim results for OAEI 2008
The growing importance of ontology mapping on the Semantic Webhas highlighted the need to manage the uncertain nature of interpreting semanticmeta data represented by heterogeneous ontologies. Considering this uncertaintyone can potentially improve the ontology mapping precision, which can lead tobetter acceptance of systems that operate in this environment. Further the applicationof different techniques like computational linguistics or belief conflictresolution that can contribute the development of better mapping algorithms arerequired in order to process the incomplete and inconsistent information used andproduced during any mapping algorithm. In this paper we introduce our algorithmcalled “DSSim” and describe the improvements that we have made compared toOAEI 2006 and OAEI 2007.
The growing importance of ontology mapping on the Semantic Webhas highlighted the need to manage the uncertain nature of interpreting semanticmeta data represented by heterogeneous ontologies. Considering this uncertaintyone can potentially improve the ontology mapping precision, which can lead tobetter acceptance of systems that operate in this environment. Further the applicationof different techniques like computational linguistics or belief conflictresolution that can contribute the development of better mapping algorithms arerequired in order to process the incomplete and inconsistent information used andproduced during any mapping algorithm. In this paper we introduce our algorithmcalled “DSSim” and describe the improvements that we have made compared toOAEI 2006 and OAEI 2007.
Harith Alani
Harith Alani
Harith Alani
Current state-of-the-art ontology-alignment evaluation methods arebased on the assumption that alignment relations come in two flavors: correctand incorrect. Some alignment systems find more correct mappings than othersand hence, by this assumption, they perform better. In practical applications however,it does not only matter how many correct mappings you find, but also whichcorrect mappings you find. This means that, apart from correctness, relevanceshould also be included in the evaluation procedure. In this paper we expand thesample-based evaluation of the OAEI 2007 food task with a sample evaluationthat uses relevance to prototypical search tasks as a selection criterion for thedrawing of sample mappings.
Relevance-based evaluation of alignment approaches: the OAEI 2007 food task revisited
Current state-of-the-art ontology-alignment evaluation methods arebased on the assumption that alignment relations come in two flavors: correctand incorrect. Some alignment systems find more correct mappings than othersand hence, by this assumption, they perform better. In practical applications however,it does not only matter how many correct mappings you find, but also whichcorrect mappings you find. This means that, apart from correctness, relevanceshould also be included in the evaluation procedure. In this paper we expand thesample-based evaluation of the OAEI 2007 food task with a sample evaluationthat uses relevance to prototypical search tasks as a selection criterion for thedrawing of sample mappings.
Relevance-based evaluation of alignment approaches: the OAEI 2007 food task revisited
Relevance-based evaluation of alignment approaches: the OAEI 2007 food task revisited
Paul Smart
Paul Smart
Paul Smart
Stanford University
Stanford University
Stanford University
Matching ontologies for emergency evacuation plans
Matching ontologies for emergency evacuation plans
In case of emergency, the coordination of different servicesdeals with different working methods, different languages, different instruments, different sensors and different data representations. Thus,the coordination of services includes heterogeneity problems that canbe managed with the help of ontology matching techniques. In this paper we present a scenario where the requirements for ontology matchingarise from emergency evacuation plans, in the specific domain of civilprotection applications. We envisage what kind of smart sensor technologies could be used to support critical decisions when heterogeneoussources of information have to be matched.
Matching ontologies for emergency evacuation plans
In case of emergency, the coordination of different servicesdeals with different working methods, different languages, different instruments, different sensors and different data representations. Thus,the coordination of services includes heterogeneity problems that canbe managed with the help of ontology matching techniques. In this paper we present a scenario where the requirements for ontology matchingarise from emergency evacuation plans, in the specific domain of civilprotection applications. We envisage what kind of smart sensor technologies could be used to support critical decisions when heterogeneoussources of information have to be matched.
University of Alberta
University of Alberta
University of Alberta
GeRoMeSuite is a generic model management system which providesseveral functions for managing complex data models, such as schema integration,definition and execution of schema mappings, model transformation, and matching.The system use the generic metamodel GeRoMe for representing models,and because of this, it is able to deal with models in various modeling languagessuch as XML Schema, OWL, ER, and relational schemas.A matching component for matching and alignment of schemas is also part ofthe system. We participated this year the first time in an OAEI contest in orderto evaluate and compare the performance of our matcher component with othersystems. Therefore, we focused our efforts on the ‘benchmark’ track.
Results of GeRoMeSuite for OAEI 2008
GeRoMeSuite is a generic model management system which providesseveral functions for managing complex data models, such as schema integration,definition and execution of schema mappings, model transformation, and matching.The system use the generic metamodel GeRoMe for representing models,and because of this, it is able to deal with models in various modeling languagessuch as XML Schema, OWL, ER, and relational schemas.A matching component for matching and alignment of schemas is also part ofthe system. We participated this year the first time in an OAEI contest in orderto evaluate and compare the performance of our matcher component with othersystems. Therefore, we focused our efforts on the ‘benchmark’ track.
Results of GeRoMeSuite for OAEI 2008
Results of GeRoMeSuite for OAEI 2008
David Kensche
David Kensche
David Kensche
Monika Lanzenberger
Monika Lanzenberger
Monika Lanzenberger
In this paper we built on top of recent effort in the areas of semanticsand interoperability to establish the basis for a comprehensive and sustainableapproach to the development and later management of bridging systems amonga variety of corporate system that need to be interconnected without beingindividually modified. In particular, we collect some preliminary evidence thata sustainable approach exists to the definition of mappings which can withstandchanges of the underlying classification schemes. This in turn adds evidencetowards the feasibility of a dynamic interoperable infrastructure supporting aglobal adaptive electronic market place.
Ontological mappings of product catalogues
Ontological mappings of product catalogues
In this paper we built on top of recent effort in the areas of semanticsand interoperability to establish the basis for a comprehensive and sustainableapproach to the development and later management of bridging systems amonga variety of corporate system that need to be interconnected without beingindividually modified. In particular, we collect some preliminary evidence thata sustainable approach exists to the definition of mappings which can withstandchanges of the underlying classification schemes. This in turn adds evidencetowards the feasibility of a dynamic interoperable infrastructure supporting aglobal adaptive electronic market place.
Ontological mappings of product catalogues
Ryutaro Ichise
Ryutaro Ichise
Ryutaro Ichise
This paper presents the alignment results of Lily for the ontologyalignment contest OAEI 2008. Lily is an ontology mapping system, and it hasfour main features: generic ontology matching, large scale ontology matching,semantic ontology matching and mapping debugging. In the past year, Lily hasbeen improved greatly for both function and performance. Lily submited theresults for seven alignment tasks: benchmark, anatomy, fao, directory,mldirectory, library and conference. The results of the benchmark task arepresented. The paper also discusses the strengths and weaknesses of our system.
Lily: ontology alignment results for OAEI 2008
Lily: ontology alignment results for OAEI 2008
Lily: ontology alignment results for OAEI 2008
This paper presents the alignment results of Lily for the ontologyalignment contest OAEI 2008. Lily is an ontology mapping system, and it hasfour main features: generic ontology matching, large scale ontology matching,semantic ontology matching and mapping debugging. In the past year, Lily hasbeen improved greatly for both function and performance. Lily submited theresults for seven alignment tasks: benchmark, anatomy, fao, directory,mldirectory, library and conference. The results of the benchmark task arepresented. The paper also discusses the strengths and weaknesses of our system.
Valentina Presutti
Valentina
Presutti
SAP
SAP
SAP
Peter Geibel
Peter Geibel
Peter Geibel
Towards dialogue-based interactive semantic mediation in the medical domain
We think of ontology matching as a dialogue-based interactivemediation process for which we propose a three stage model. Apreliminary evaluation shows how we applied this method of eliciting inputfor ontology matching in the medical domain. Especially, we addressthe challenge how to use dialogue-based interactivity with the user torate partial alignments between two ontologies.
Towards dialogue-based interactive semantic mediation in the medical domain
Towards dialogue-based interactive semantic mediation in the medical domain
We think of ontology matching as a dialogue-based interactivemediation process for which we propose a three stage model. Apreliminary evaluation shows how we applied this method of eliciting inputfor ontology matching in the medical domain. Especially, we addressthe challenge how to use dialogue-based interactivity with the user torate partial alignments between two ontologies.
MapPSO results for OAEI 2008
MapPSO results for OAEI 2008
MapPSO results for OAEI 2008
We present first results of an ontology alignment approach that isbased on discrete particle swarm optimisation. Firstly in this paper we will describe,how the algorithm approaches the ontology matching task as an optimisationproblem, and briefly sketch how the specific technique of particle swarmoptimisation is applied. Secondly, we will briefly discuss the results gained forthe Benchmark data set of the 2008 Ontology Alignment Evaluation Intiative.
We present first results of an ontology alignment approach that isbased on discrete particle swarm optimisation. Firstly in this paper we will describe,how the algorithm approaches the ontology matching task as an optimisationproblem, and briefly sketch how the specific technique of particle swarmoptimisation is applied. Secondly, we will briefly discuss the results gained forthe Benchmark data set of the 2008 Ontology Alignment Evaluation Intiative.
Ricoh Europe plc.
Ricoh Europe plc.
Ricoh Europe plc.
Enrico Motta
Enrico Motta
Enrico Motta
Peter Mork
Peter Mork
Peter Mork
ISTI-C.N.R.
ISTI-C.N.R.
ISTI-C.N.R.
Miklos Nagy
Miklos Nagy
Miklos Nagy
University of Mannheim
University of Mannheim
University of Mannheim
US National Library of Medicine
US National Library of Medicine
US National Library of Medicine
In this report, we give a brief explanation of how RiMOM obtains the ontology alignment results at OAEI 2008 contest. We introduce the alignment process of RiMOM and more than 8 different alignment strategies integrated in RiMOM. Since every strategy is defined based on one specific ontological-information, we, in particular, study how the different strategies perform for different alignment tasks in the contest and design a strategy selection technique to get better performance. The result shows this technique is very useful. We also discuss some future work about RiMOM.
In this report, we give a brief explanation of how RiMOM obtains the ontology alignment results at OAEI 2008 contest. We introduce the alignment process of RiMOM and more than 8 different alignment strategies integrated in RiMOM. Since every strategy is defined based on one specific ontological-information, we, in particular, study how the different strategies perform for different alignment tasks in the contest and design a strategy selection technique to get better performance. The result shows this technique is very useful. We also discuss some future work about RiMOM.
RiMOM results for OAEI 2008
RiMOM results for OAEI 2008
RiMOM results for OAEI 2008
Towards ontology interoperability through conceptual groundings
The widespread use of ontologies raises the need to resolve heterogeneities betweendistinct conceptualisations in order to support interoperability. The aim of ontology mapping is,to establish formal relations between a set of knowledge entities which represent the same or asimilar meaning in distinct ontologies. Whereas the symbolic approach of established SWrepresentation standards – based on first-order logic and syllogistic reasoning – does notimplicitly represent similarity relationships, the ontology mapping task strongly relies onidentifying semantic similarities. However, while concept representations across distinctontologies hardly equal another, manually or even semi-automatically identifying similarityrelationships is costly. Conceptual Spaces (CS) enable the representation of concepts as vectorspaces which implicitly carry similarity information. But CS provide neither an implicitrepresentational mechanism nor a means to represent arbitrary relations between concepts orinstances. In order to overcome these issues, we propose a hybrid knowledge representationapproach which extends first-order logic ontologies with a conceptual grounding through a setof CS-based representations. Consequently, semantic similarity between instances –represented as members in CS – is indicated by means of distance metrics. Hence, automaticsimilarity-detection between instances across distinct ontologies is supported in order tofacilitate ontology mapping.
Towards ontology interoperability through conceptual groundings
The widespread use of ontologies raises the need to resolve heterogeneities betweendistinct conceptualisations in order to support interoperability. The aim of ontology mapping is,to establish formal relations between a set of knowledge entities which represent the same or asimilar meaning in distinct ontologies. Whereas the symbolic approach of established SWrepresentation standards – based on first-order logic and syllogistic reasoning – does notimplicitly represent similarity relationships, the ontology mapping task strongly relies onidentifying semantic similarities. However, while concept representations across distinctontologies hardly equal another, manually or even semi-automatically identifying similarityrelationships is costly. Conceptual Spaces (CS) enable the representation of concepts as vectorspaces which implicitly carry similarity information. But CS provide neither an implicitrepresentational mechanism nor a means to represent arbitrary relations between concepts orinstances. In order to overcome these issues, we propose a hybrid knowledge representationapproach which extends first-order logic ontologies with a conceptual grounding through a setof CS-based representations. Consequently, semantic similarity between instances –represented as members in CS – is indicated by means of distance metrics. Hence, automaticsimilarity-detection between instances across distinct ontologies is supported in order tofacilitate ontology mapping.
Towards ontology interoperability through conceptual groundings
Bosch Research
Bosch Research
Bosch Research
Piotr Stolarski
Piotr Stolarski
Piotr Stolarski
Paulo Quaresma
Paulo Quaresma
Paulo Quaresma
This article describes a base system for ontology alignment, SAMBO,and an extension, SAMBOdtf.We present their results for the benchmark, anatomyand FAO tasks in the 2008 Ontology Alignment Evaluation Initiative. For thebenchmark and FAO tasks SAMBO uses a strategy based on string matching aswell as the use of a thesaurus. It obtains good results in many cases. For theanatomy task SAMBO uses a combination of string matching and the use of domainknowledge. This combination performed well in former evaluations usingother anatomy ontologies. SAMBOdtf uses the same strategies but, in addition,uses an advanced filtering technique that augments recall while maintaining ahigh precision.
SAMBO and SAMBOdtf results for the Ontology Alignment Evaluation Initiative 2008
SAMBO and SAMBOdtf results for the Ontology Alignment Evaluation Initiative 2008
SAMBO and SAMBOdtf results for the Ontology Alignment Evaluation Initiative 2008
This article describes a base system for ontology alignment, SAMBO,and an extension, SAMBOdtf.We present their results for the benchmark, anatomyand FAO tasks in the 2008 Ontology Alignment Evaluation Initiative. For thebenchmark and FAO tasks SAMBO uses a strategy based on string matching aswell as the use of a thesaurus. It obtains good results in many cases. For theanatomy task SAMBO uses a combination of string matching and the use of domainknowledge. This combination performed well in former evaluations usingother anatomy ontologies. SAMBOdtf uses the same strategies but, in addition,uses an advanced filtering technique that augments recall while maintaining ahigh precision.
Xiao Zhang
Xiao Zhang
Xiao Zhang
Md. Hanif Seddiqui
Md. Hanif Seddiqui
Md. Hanif Seddiqui
Christian Meilicke
Christian Meilicke
Christian Meilicke
Informatica Trentina S.P.A.
Informatica Trentina S.P.A.
Informatica Trentina S.P.A.
Spider: bringing non-equivalence mappings to OAEI
With the large majority of existing matching systems focusing on derivingequivalence mappings, OAEI has been primarily focused on assessing suchkind of relations. As the field inevitably advances towards the discovery of morecomplex mappings, the contest will need to reflect such changes as well. In thispaper we present Spider, a system that provides alignments containing not onlyequivalence mappings, but also a variety of different mapping types (namely,subsumption, disjointness and named relations). Our goal is both to get an insightinto the functioning of our system and, more importantly, to assess the currentsupport for dealing with non-equivalence mappings in the OAEI contest. Wehope that our observations will contribute to further enhance the procedure of thecontest.
Spider: bringing non-equivalence mappings to OAEI
Spider: bringing non-equivalence mappings to OAEI
With the large majority of existing matching systems focusing on derivingequivalence mappings, OAEI has been primarily focused on assessing suchkind of relations. As the field inevitably advances towards the discovery of morecomplex mappings, the contest will need to reflect such changes as well. In thispaper we present Spider, a system that provides alignments containing not onlyequivalence mappings, but also a variety of different mapping types (namely,subsumption, disjointness and named relations). Our goal is both to get an insightinto the functioning of our system and, more importantly, to assess the currentsupport for dealing with non-equivalence mappings in the OAEI contest. Wehope that our observations will contribute to further enhance the procedure of thecontest.
Yannis Kalfoglou
Yannis Kalfoglou
Yannis Kalfoglou
Paolo Besana
Paolo Besana
Paolo Besana
Baowen Xu
Baowen Xu
Baowen Xu
Masaki Aono
Masaki Aono
Masaki Aono
Tassos Venetis
Tassos Venetis
Tassos Venetis
TaxoMap is an alignment tool which aim is to discover rich correspondencesbetween concepts. It performs an oriented alignment (from a sourceto a target ontology) and takes into account labels and sub-class descriptions.Our participation in last year edition of the competition have put the emphasis oncertain limits. TaxoMap 2 is a new implementation of TaxoMap that reduces significantlyruntime and enables parameterization by specifying the ontology languageand different thresholds used to extract different mapping relations. Thenew implementation stresses on terminological techniques, it takes into accountsynonymy, and multi-label description of concepts. Special effort was made tohandle large-scale ontologies by partitioning input ontologies into modules toalign. We conclude the paper by pointing out the necessary improvements thatneed to be made.
TaxoMap in the OAEI 2008 alignment contest
TaxoMap in the OAEI 2008 alignment contest
TaxoMap is an alignment tool which aim is to discover rich correspondencesbetween concepts. It performs an oriented alignment (from a sourceto a target ontology) and takes into account labels and sub-class descriptions.Our participation in last year edition of the competition have put the emphasis oncertain limits. TaxoMap 2 is a new implementation of TaxoMap that reduces significantlyruntime and enables parameterization by specifying the ontology languageand different thresholds used to extract different mapping relations. Thenew implementation stresses on terminological techniques, it takes into accountsynonymy, and multi-label description of concepts. Special effort was made tohandle large-scale ontologies by partitioning input ontologies into modules toalign. We conclude the paper by pointing out the necessary improvements thatneed to be made.
TaxoMap in the OAEI 2008 alignment contest
Qiang Liu
Qiang Liu
Qiang Liu
Ondrej Svab-Zamazal
Ondrej Svab-Zamazal
Ondrej Svab-Zamazal
Cassia Trojahn
Cassia Trojahn
Cassia Trojahn
University of Calabria
University of Calabria
University of Calabria
Aldo Gangemi
Aldo
Gangemi
Marta Sabou
Marta Sabou
Marta Sabou
David Macii
David Macii
David Macii
INRIA
INRIA
INRIA
Clinical Informatics R&D, Partners HealthCare System
Clinical Informatics R&D, Partners HealthCare System
Clinical Informatics R&D, Partners HealthCare System
Yves R. Jean-Mary
Yves R. Jean-Mary
Yves R. Jean-Mary
Caterina Caracciolo
Caterina Caracciolo
Caterina Caracciolo
Haifa Zargayouna
Haifa Zargayouna
Haifa Zargayouna
He Tan
He Tan
He Tan
Yannis Velegrakis
Yannis Velegrakis
Yannis Velegrakis
Eleni Stroulia
Eleni Stroulia
Eleni Stroulia
IIIA-CSIC
IIIA-CSIC
IIIA-CSIC
TasLab
TasLab
TasLab
Ivan Pilati
Ivan Pilati
Ivan Pilati
University of Illinois at Chicago
University of Illinois at Chicago
University of Illinois at Chicago
Vassilis Tzouvaras
Vassilis Tzouvaras
Vassilis Tzouvaras
Peng Wang
Peng Wang
Peng Wang
DFKI
DFKI
DFKI
Qian Zhong
Qian Zhong
Qian Zhong
Veronique Malaise
Veronique Malaise
Veronique Malaise
National Institute of Informatics
National Institute of Informatics
National Institute of Informatics
On applying matching tools to large-scale ontologies
Many existing ontology matching tools are not well scalable.In this paper, we present the Malasco system, which serves as a framework for reusing existing, non-scalable matching systems on large-scaleontologies. The results achieved with different combinations of partitioning and matching tools are discussed, and optimization techniques areexamined. It is shown that the loss of result quality when matching withpartitioned data can be reduced to less than 5% compared to matchingwith unpartitioned data.
On applying matching tools to large-scale ontologies
On applying matching tools to large-scale ontologies
Many existing ontology matching tools are not well scalable.In this paper, we present the Malasco system, which serves as a framework for reusing existing, non-scalable matching systems on large-scaleontologies. The results achieved with different combinations of partitioning and matching tools are discussed, and optimization techniques areexamined. It is shown that the loss of result quality when matching withpartitioned data can be reduced to less than 5% compared to matchingwith unpartitioned data.
Jorge Gracia
Jorge Gracia
Jorge Gracia
Isabel Cruz
Isabel Cruz
Isabel Cruz
Eduardo Mena
Eduardo Mena
Eduardo Mena
Literature-based alignment of ontologies
Literature-based alignment of ontologies
Literature-based alignment of ontologies
In this paper we propose and evaluate new strategies foraligning ontologies based on text categorization of literature using supportvector machines-based text classifiers, and compare them with existingliterature-based strategies. We also compare and combine thesestrategies with linguistic strategies.
In this paper we propose and evaluate new strategies foraligning ontologies based on text categorization of literature using supportvector machines-based text classifiers, and compare them with existingliterature-based strategies. We also compare and combine thesestrategies with linguistic strategies.
Ontology mapping via structural and instance-based similarity measures
The paper presents an overview of a novel procedure formapping hierarchical ontologies, populated with properly classified textdocuments. It combines structural and instance-based approaches to reducethe terminological and conceptual ontology heterogeneity. It yieldsimportant granularity and instantiation judgments about the inputs andis to be applied to mapping web-directories.
Ontology mapping via structural and instance-based similarity measures
Ontology mapping via structural and instance-based similarity measures
The paper presents an overview of a novel procedure formapping hierarchical ontologies, populated with properly classified textdocuments. It combines structural and instance-based approaches to reducethe terminological and conceptual ontology heterogeneity. It yieldsimportant granularity and instantiation judgments about the inputs andis to be applied to mapping web-directories.
Sandra Geisler
Sandra Geisler
Sandra Geisler
Daniele Montanari
Daniele Montanari
Daniele Montanari
Guus Schreiber
Guus Schreiber
Guus Schreiber
The Automated Semantic Mapping of Ontologies with Validation (ASMOV) algorithm for ontology alignment was one of the top performing algorithms in the 2007 Ontology Alignment Evaluation Initiative (OAEI). In this paper, we present a brief overview of the algorithm and its improvements, followed by an analysis of its results on the 2008 OAEI tests.
ASMOV: results for OAEI 2008
ASMOV: results for OAEI 2008
The Automated Semantic Mapping of Ontologies with Validation (ASMOV) algorithm for ontology alignment was one of the top performing algorithms in the 2007 Ontology Alignment Evaluation Initiative (OAEI). In this paper, we present a brief overview of the algorithm and its improvements, followed by an analysis of its results on the 2008 OAEI tests.
ASMOV: results for OAEI 2008
Testing the impact of pattern-based ontology refactoring on ontology matching results
We observe the impact of ontology refactoring, based on detectionof name patterns in the ontology structure, on the results of ontology matching.Results of our experiment are evaluated using novel logic-based measures accompaniedby an analysis of typical effects. Although the pattern detection methodonly covers a fraction of ontological errors, there seems to be a measurable effecton the quality of the resulting matching.
Testing the impact of pattern-based ontology refactoring on ontology matching results
Testing the impact of pattern-based ontology refactoring on ontology matching results
We observe the impact of ontology refactoring, based on detectionof name patterns in the ontology structure, on the results of ontology matching.Results of our experiment are evaluated using novel logic-based measures accompaniedby an analysis of typical effects. Although the pattern detection methodonly covers a fraction of ontological errors, there seems to be a measurable effecton the quality of the resulting matching.
Heiko Paulheim
Heiko Paulheim
Heiko Paulheim
Ontology matching with CIDER: evaluation report for the OAEI 2008
Ontology matching, the task of determining relations thathold among terms of two different ontologies, is a key issue in the Semantic Web and other related fields. In order to compare the behaviour of different ontology matching systems, the Ontology Alignment Evaluation Initiative (OAEI) has established a periodical controlled evaluationthat comes in a yearly event. We present here our participation in the2008 initiative.Our schema-based alignment algorithm compares each pair of ontologyterms by, firstly, extracting their ontological contexts up to a certaindepth (enriched by using transitive entailment) and, secondly, combining different elementary ontology matching techniques (e.g., lexical distances and vector space modelling). Benchmark results show a very goodbehaviour in terms of precision, while preserving an acceptable recall.Based on our experience, we have also included some remarks about thenature of benchmark test cases that, in our opinion, could help improvingthe OAEI tests in the future.
Ontology matching, the task of determining relations thathold among terms of two different ontologies, is a key issue in the Semantic Web and other related fields. In order to compare the behaviour of different ontology matching systems, the Ontology Alignment Evaluation Initiative (OAEI) has established a periodical controlled evaluationthat comes in a yearly event. We present here our participation in the2008 initiative.Our schema-based alignment algorithm compares each pair of ontologyterms by, firstly, extracting their ontological contexts up to a certaindepth (enriched by using transitive entailment) and, secondly, combining different elementary ontology matching techniques (e.g., lexical distances and vector space modelling). Benchmark results show a very goodbehaviour in terms of precision, while preserving an acceptable recall.Based on our experience, we have also included some remarks about thenature of benchmark test cases that, in our opinion, could help improvingthe OAEI tests in the future.
Ontology matching with CIDER: evaluation report for the OAEI 2008
Ontology matching with CIDER: evaluation report for the OAEI 2008
Ludger van Elst
Ludger van Elst
Ludger van Elst
Xiang Li
Xiang Li
Xiang Li
Stefan Dietze
Stefan Dietze
Stefan Dietze
Southeast University
Southeast University
Southeast University
Giorgos Stoilos
Giorgos Stoilos
Giorgos Stoilos
Olivier Bodenreider
Olivier Bodenreider
Olivier Bodenreider
Fausto Giunchiglia
Fausto Giunchiglia
Fausto Giunchiglia
ad1bf97094dfdc9d370fdf06df968c90c4b4852d
Jan Hettenhausen
Jan Hettenhausen
Jan Hettenhausen
John Domingue
John Domingue
John Domingue
AROMA results for OAEI 2008
AROMA results for OAEI 2008
AROMA results for OAEI 2008
This paper presents the results obtained by AROMA for its first participationto OAEI. AROMA is an hybrid, extensional and asymmetric ontologyalignment method which makes use of the association paradigm and a statisticalinterstingness measure, the implication intensity.
This paper presents the results obtained by AROMA for its first participationto OAEI. AROMA is an hybrid, extensional and asymmetric ontologyalignment method which makes use of the association paradigm and a statisticalinterstingness measure, the implication intensity.
Linköpings Universitet
Linköpings Universitet
Linköpings Universitet
Wei Xu
Wei Xu
Wei Xu
Wei Hu
Wei Hu
Wei Hu
Alignment results of Anchor-Flood algorithm for OAEI-2008
Our proposed algorithm called Anchor-Flood algorithm, starts off withanchors. It gradually explores concepts by collecting neighbors in concept taxonomy,thereby taking advantage of locality of reference in the graph data structure.Then local alignment process runs over the collected small blocks of concepts.The process is repeated for the newly found aligned pairs. In this way, we cansignificantly reduce the computational time for the alignment as our algorithmconcentrates on the aligned pairs and it resolves the scalability problem in ontologyalignment over large ontologies. Through several experiments against OAEI-2008 datasets, we will demonstrate the results and the features of our Anchor-Flood algorithm.
Alignment results of Anchor-Flood algorithm for OAEI-2008
Our proposed algorithm called Anchor-Flood algorithm, starts off withanchors. It gradually explores concepts by collecting neighbors in concept taxonomy,thereby taking advantage of locality of reference in the graph data structure.Then local alignment process runs over the collected small blocks of concepts.The process is repeated for the newly found aligned pairs. In this way, we cansignificantly reduce the computational time for the alignment as our algorithmconcentrates on the aligned pairs and it resolves the scalability problem in ontologyalignment over large ontologies. Through several experiments against OAEI-2008 datasets, we will demonstrate the results and the features of our Anchor-Flood algorithm.
Alignment results of Anchor-Flood algorithm for OAEI-2008
Jie Tang
Jie Tang
Jie Tang
Davide Lorusso
Davide Lorusso
Davide Lorusso
Maria Vargas-Vera
Maria Vargas-Vera
Maria Vargas-Vera
First results of the Ontology Alignment Evaluation Initiative 2008
First results of the Ontology Alignment Evaluation Initiative 2008
Ontology matching consists of finding correspondences between ontologyentities. OAEI campaigns aim at comparing ontology matching systems onprecisely defined test sets. Test sets can use ontologies of different nature (fromexpressive OWL ontologies to simple directories) and use different modalities(e.g., blind evaluation, open evaluation, consensus). OAEI-2008 builds over previouscampaigns by having 4 tracks with 8 test sets followed by 13 participants.Following the trend of previous years, more participants evolve at the forefront.The final and official results of the campaign are those published on the OAEIweb site.
Ontology matching consists of finding correspondences between ontologyentities. OAEI campaigns aim at comparing ontology matching systems onprecisely defined test sets. Test sets can use ontologies of different nature (fromexpressive OWL ontologies to simple directories) and use different modalities(e.g., blind evaluation, open evaluation, consensus). OAEI-2008 builds over previouscampaigns by having 4 tracks with 8 test sets followed by 13 participants.Following the trend of previous years, more participants evolve at the forefront.The final and official results of the campaign are those published on the OAEIweb site.
First results of the Ontology Alignment Evaluation Initiative 2008
Baoshi Yan
Baoshi Yan
Baoshi Yan
A community based approach for managing ontology alignments
The Semantic Web is rapidly becoming a defacto distributedrepository for semantically represented data, thus leveraging on the addedon value of the network effect. Various ontology mapping techniques andtools have been devised to facilitate the bridging and integration of distributeddata repositories. Nevertheless, ontology mapping can benefitfrom human supervision to increase accuracy of results. The spread ofWeb 2.0 approaches demonstrate the possibility of using collaborativetechniques for reaching consensus. While a number of prototypes for collaborativeontology construction are being developed, collaborative ontologymapping is not yet well investigated. In this paper, we describe aprototype that combines off-the-shelf ontology mapping tools with socialsoftware techniques to enable users to collaborate on mapping ontologies.
A community based approach for managing ontology alignments
The Semantic Web is rapidly becoming a defacto distributedrepository for semantically represented data, thus leveraging on the addedon value of the network effect. Various ontology mapping techniques andtools have been devised to facilitate the bridging and integration of distributeddata repositories. Nevertheless, ontology mapping can benefitfrom human supervision to increase accuracy of results. The spread ofWeb 2.0 approaches demonstrate the possibility of using collaborativetechniques for reaching consensus. While a number of prototypes for collaborativeontology construction are being developed, collaborative ontologymapping is not yet well investigated. In this paper, we describe aprototype that combines off-the-shelf ontology mapping tools with socialsoftware techniques to enable users to collaborate on mapping ontologies.
A community based approach for managing ontology alignments
Konstantin Todorov
Konstantin Todorov
Konstantin Todorov
Shenghui Wang
Shenghui Wang
Shenghui Wang
State-of-the art mappers articulate several techniques usingdifferent sources of knowledge in an unified process. An important issueof ontology mapping is to find ways of choosing among many techniquesand their variations, and then combining their results. For this,an innovative and promising option is to use frameworks dealing witharguments for or against correspondences. In this paper, we re-use anargumentation framework that considers the confidence levels of mappingarguments. We also propose new frameworks that use voting as away to cope with various degrees of consensus among arguments. Wecompare these frameworks by evaluating their application to a range ofindividual mappers, in the context of a real-world library case.
Using quantitative aspects of alignment generation for argumentation on mappings
Using quantitative aspects of alignment generation for argumentation on mappings
State-of-the art mappers articulate several techniques usingdifferent sources of knowledge in an unified process. An important issueof ontology mapping is to find ways of choosing among many techniquesand their variations, and then combining their results. For this,an innovative and promising option is to use frameworks dealing witharguments for or against correspondences. In this paper, we re-use anargumentation framework that considers the confidence levels of mappingarguments. We also propose new frameworks that use voting as away to cope with various degrees of consensus among arguments. Wecompare these frameworks by evaluating their application to a range ofindividual mappers, in the context of a real-world library case.
Using quantitative aspects of alignment generation for argumentation on mappings
University of Trento
University of Trento
University of Trento
Natasha Noy
Natasha Noy
Natasha Noy
Faycal Hamdi
Faycal Hamdi
Faycal Hamdi
Alfio Ferrara
Alfio Ferrara
Alfio Ferrara
Anna Lisa Gentile
Anna Lisa
Gentile
The MITRE Corporation
The MITRE Corporation
The MITRE Corporation
Andrea Giovanni Nuzzolese
Andrea Giovanni
Nuzzolese
On fixing semantic alignment evaluation measures
On fixing semantic alignment evaluation measures
On fixing semantic alignment evaluation measures
The evaluation of ontology matching algorithms mainly consists ofcomparing a produced alignment with a reference one. Usually, this evaluationrelies on the classical precision and recall measures. This evaluation model is notsatisfactory since it does not take into account neither the closeness of correspondances,nor the semantics of alignments. A first solution consists of generalizingthe precision and recall measures in order to solve the problem of rigidity of classicalmodel. Another solution aims at taking advantage of the semantic of alignmentsin the evaluation. In this paper, we show and analyze the limits of theseevaluation models. Given that measures values depend on the syntactic form ofthe alignment, we first propose an normalization of alignment. Then, we proposetwo new sets of evaluation measures. The first one is a semantic extension of relaxedprecision and recall. The second one consists of bounding the alignmentspace to make ideal semantic precision and recall applicable.
The evaluation of ontology matching algorithms mainly consists ofcomparing a produced alignment with a reference one. Usually, this evaluationrelies on the classical precision and recall measures. This evaluation model is notsatisfactory since it does not take into account neither the closeness of correspondances,nor the semantics of alignments. A first solution consists of generalizingthe precision and recall measures in order to solve the problem of rigidity of classicalmodel. Another solution aims at taking advantage of the semantic of alignmentsin the evaluation. In this paper, we show and analyze the limits of theseevaluation models. Given that measures values depend on the syntactic form ofthe alignment, we first propose an normalization of alignment. Then, we proposetwo new sets of evaluation measures. The first one is a semantic extension of relaxedprecision and recall. The second one consists of bounding the alignmentspace to make ideal semantic precision and recall applicable.
Traditionally, the quality of ontology matching is measured using precisionand recall with respect to a reference mapping. These measures have atleast two major drawbacks. First, a mapping with acceptable precision and recallmight nevertheless suffer from internal logical problems that hinder a sensible useof the mapping. Second, in practical situations reference mappings are not available.To avoid these drawbacks we introduce quality measures that are based onthe notion of mapping incoherence that can be used without a reference mapping.We argue that these measures are a reasonable complement to the well-knownmeasures already used for mapping evaluation. In particular, we show that one ofthese measures provides a strict upper bound for the precision of a mapping.
Incoherence as a basis for measuring the quality of ontology mappings
Incoherence as a basis for measuring the quality of ontology mappings
Incoherence as a basis for measuring the quality of ontology mappings
Traditionally, the quality of ontology matching is measured using precisionand recall with respect to a reference mapping. These measures have atleast two major drawbacks. First, a mapping with acceptable precision and recallmight nevertheless suffer from internal logical problems that hinder a sensible useof the mapping. Second, in practical situations reference mappings are not available.To avoid these drawbacks we introduce quality measures that are based onthe notion of mapping incoherence that can be used without a reference mapping.We argue that these measures are a reasonable complement to the well-knownmeasures already used for mapping evaluation. In particular, we show that one ofthese measures provides a strict upper bound for the precision of a mapping.
Willem Robert van Hage
Willem Robert van Hage
Willem Robert van Hage
Jérôme Euzenat
Jérôme Euzenat
Jerome Euzenat
Jerome Euzenat
22bc9bfef9e568c1439538b6a03d4cd8becf01f9
Jerome Euzenat
Jérôme Euzenat
Resolution of conicts among ontology mappings: a fuzzy approach
Interoperability is a strong requirement in open distributedsystems and in the Semantic Web. The need for ontology integrationis not always completely met by the available ontology matching techniques because, in most cases, the semantics of the compared ontologiesis not considered, thus leading to inconsistent mappings. Probabilisticapproaches has been proposed to validate mappings and solve the inconsistencies, based on a mapping confidence measure. As probabilisticapproaches suffer from the lack of well-founded likelihood measures ofmapping correctness, we propose a validation approach based on fuzzyinterpretation of mappings, which better models the notion of degreeof similarity between ontology elements. Moreover, we describe a conflict resolution method which computes the minimal sets of conflictingmappings and can be the ground of different validation strategies.
Resolution of conicts among ontology mappings: a fuzzy approach
Resolution of conicts among ontology mappings: a fuzzy approach
Interoperability is a strong requirement in open distributedsystems and in the Semantic Web. The need for ontology integrationis not always completely met by the available ontology matching techniques because, in most cases, the semantics of the compared ontologiesis not considered, thus leading to inconsistent mappings. Probabilisticapproaches has been proposed to validate mappings and solve the inconsistencies, based on a mapping confidence measure. As probabilisticapproaches suffer from the lack of well-founded likelihood measures ofmapping correctness, we propose a validation approach based on fuzzyinterpretation of mappings, which better models the notion of degreeof similarity between ontology elements. Moreover, we describe a conflict resolution method which computes the minimal sets of conflictingmappings and can be the ground of different validation strategies.
Anthony Jameson
Anthony Jameson
Anthony Jameson
Luciano Serafini
Luciano Serafini
Luciano Serafini
Mansur R. Kabuka
Mansur R. Kabuka
Mansur R. Kabuka
Daniel Sonntag
Daniel Sonntag
Daniel Sonntag
Marco Schorlemmer
Marco Schorlemmer
Marco Schorlemmer
Gianluca Correndo
Gianluca Correndo
Gianluca Correndo
Vojtech Svatek
Vojtech Svatek
Vojtech Svatek
Jerome David
Jerome David
Jerome David
York Sure
York Sure
York Sure
Paolo Bouquet
Paolo Bouquet
Paolo Bouquet
Hap Kolb
Hap Kolb
Hap Kolb
Luigi Palopoli
Luigi Palopoli
Luigi Palopoli
Juan Pane
Juan Pane
Juan Pane
Antoine Isaac
Antoine Isaac
Antoine Isaac
Pavel Shvaiko
Pavel Shvaiko
Pavel Shvaiko
6fc18ceccfe31ac7c9d76fea4e28a96b6fdd804e
Juanzi Li
Juanzi Li
Juanzi Li
Laura Hollink
Laura Hollink
Laura Hollink
Fabio Andreatta
Fabio Andreatta
Fabio Andreatta
Chantal Reynaud
Chantal Reynaud
Chantal Reynaud
Giorgos Stamou
Giorgos Stamou
Giorgos Stamou
Jurgen Bock
Jurgen Bock
Jurgen Bock
Vienna University of Technology
Vienna University of Technology
Vienna University of Technology
University of Edinburgh
University of Edinburgh
University of Edinburgh
Heiner Stuckenschmidt
f072571d8c8abd21e637a9dec2c5e835165c8ece
Heiner Stuckenschmidt
Heiner Stuckenschmidt
Domenico Beneventano
Domenico Beneventano
Domenico Beneventano
Patrick Lambrix
Patrick Lambrix
Patrick Lambrix
FBK-IRST
FBK-IRST
FBK-IRST
Brigitte Safar
Brigitte Safar
Brigitte Safar
Umberto Straccia
Umberto Straccia
Umberto Straccia
Chinese Academy of Sciences
Chinese Academy of Sciences
Chinese Academy of Sciences
LIG
LIG
LIG
Vrije Universiteit Amsterdam
Vrije Universiteit Amsterdam
Vrije Universiteit Amsterdam
Luca Mion
Luca Mion
Luca Mion
Christoph Quix
Christoph Quix
Christoph Quix
Vipul Kashyap
Vipul Kashyap
Vipul Kashyap
Songmao Zhang
Songmao Zhang
Songmao Zhang