Cromatcher: An Ontology Matching System Based on Automated Weighted Aggregation and Iterative Final Alignment

Gulić, Marko; Vrdojak, Boris; Banek, Marko

doi:10.2139/ssrn.3199271

Cited by 11 publications

(25 citation statements)

References 42 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…CroMatch (Gulić et al, 2016), AML (Cruz et al, 2009), XMap (Djeddi andKhadir, 2010) perform ontology matching based on heuristic methods that rely on aggregation functions. LogMap and LogMapBio (Jiménez-Ruiz and Grau, 2011) use logic-based reasoning over the extracted features and cast the ontology matching to a satisfiability problem.…”

Section: Resultsmentioning

confidence: 99%

“…The vast majority of ontology matching research follows the feature engineering approach (Wang and Xu, 2008;Cruz et al, 2009;Khadir et al, 2011;Jiménez-Ruiz and Grau, 2011;Fahad et al, 2012;Ngo and Bellahsene, 2012;Gulić et al, 2016). Features are generated using a broad range of techniques (Anam et al, 2015;Harispe et al, 2015), ranging from the exploitation of terminological information, including structural similarities and logical constraints, such as datatype properties, cardinality constraints, etc.…”

Section: Selecting Features For Ontology Matchingmentioning

confidence: 99%

See 1 more Smart Citation

DeepAlignment: Unsupervised Ontology Matching with Refined Word Vectors

Kolyvakis¹,

Kalousis²,

Kiritsis³

2018

Proceedings of the 2018 Conference of the North American Chapter Of the Association for Computational Linguistics: Hu

View full text Add to dashboard Cite

Ontologies compartmentalize types and relations in a target domain and provide the semantic backbone needed for a plethora of practical applications. Very often different ontologies are developed independently for the same domain. Such "parallel" ontologies raise the need for a process that will establish alignments between their entities in order to unify and extend the existing knowledge. In this work, we present a novel entity alignment method which we dub DeepAlignment. DeepAlignment refines pre-trained word vectors aiming at deriving ontological entity descriptions which are tailored to the ontology matching task. The absence of explicit information relevant to the ontology matching task during the refinement process makes DeepAlignment completely unsupervised. We empirically evaluate our method using standard ontology matching benchmarks. We present significant performance improvements over the current state-of-the-art, demonstrating the advantages that representation learning techniques bring to ontology matching.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Selecting Features For Ontology Matchingmentioning

confidence: 99%

DeepAlignment: Unsupervised Ontology Matching with Refined Word Vectors

Kolyvakis¹,

Kalousis²,

Kiritsis³

2018

Proceedings of the 2018 Conference of the North American Chapter Of the Association for Computational Linguistics: Hu

View full text Add to dashboard Cite

show abstract

“…In this subsection, the basic terms referring to ontology matching, adopted from [4], [6], are presented.…”

Section: Terminologymentioning

confidence: 99%

“…First, we improved our final alignment method presented in [6] by introducing the automatic adjustment of correspondence threshold value instead of manual adjustment which was performed by a user of our ontology matching system. Only the correspondences that have the highest value for both ontology entities (with respect to all entities from the other ontology) and have the similarity value greater than automatically adjusted correspondence threshold enter the final alignment.…”

Section: Introductionmentioning

confidence: 99%

An Iterative Automatic Final Alignment Method in the Ontology Matching System

Gulić

Vrdoljak

Vuković

2018

JIOS

View full text Add to dashboard Cite

Ontology matching plays an important role in the integration of heterogeneous data sources that are described by ontologies. In order to determine correspondences between ontologies, a set of matchers can be used. After the execution of these matchers and the aggregation of the results obtained by these matchers, a final alignment method is executed in order to select appropriate correspondences between entities of compared ontologies. The final alignment method is an important part of the ontology matching process because it directly determines the output result of this process. In this paper we improve our iterative final alignment method by introducing an automatic adjustment of final alignment threshold as well as a new rule for determining false correspondences with similarity values greater than adjusted threshold. An evaluation of the method is performed on the test ontologies of the OAEI evaluation contest and a comparison with other final alignment methods is given.

show abstract

“…CroMatcher [34] is an automatic ontology matching system which denotes the resemblance or correlation between the entities of two different ontologies. It analyses the arrangement delivered by the matchers and highlights the one with the unique and particular arrangement.…”

Section: Ontology Matchingmentioning

confidence: 99%

A Known in Advance, What Ontologies to Integrate? For Effective Ontology Merging Using K-means Clustering

Makwana¹,

Ganatra²

2018

IJIES

View full text Add to dashboard Cite

A significant aspect of any system is how we present the knowledge. Ontology is one of the methods to present shared knowledge of the particular domain. Ontology can be designed and developed on specific domain or subdomain by many groups and researchers, which will create heterogeneity. To solve this problem ontology integration is necessary. To integrate shared knowledge, we require calculating the similarity between two ontologies, selected from a corpus using ontology matching techniques. Here we define a procedure to create a group of ontologies. Ontology comparison can be made using tools to find similar classes. As a similarity measure, Jaccard Similarity Index (JSI) is used create a group of ontology by an algorithm named k means. For each cluster, we generate buckets of ontologies, and from the bucket, all ontologies are grouped in one ontology reducing the attempt of exploring in enquiring understanding in multiple ontologies. Here we have check performance of ontology matching; clustering and merging algorithms script using standard benchmarking techniques of agriculture domain and conference track. Also, compare response time performance of SPARQL (SPARQL Protocol and RDF Query Language) query on merged ontology using this method. For experimentation, we have selected ontology corpus from agriculture domain and conference tack domain of OAEI (Ontology Alignment Evaluation Initiative). At the end of experimentation through our proposed approach we could achieve improvement in average loading and match time in ontology matching process compare to an existing tool. Also, we could achieve significance result in ontology merging process though benchmark parameter of coverage, compactness, and average merge time of two ontologies. Finally, in SPARQL query experimentation, we got success in improvement in reducing search space and response time of the query.

show abstract

Cromatcher: An Ontology Matching System Based on Automated Weighted Aggregation and Iterative Final Alignment

Cited by 11 publications

References 42 publications

DeepAlignment: Unsupervised Ontology Matching with Refined Word Vectors

DeepAlignment: Unsupervised Ontology Matching with Refined Word Vectors

An Iterative Automatic Final Alignment Method in the Ontology Matching System

A Known in Advance, What Ontologies to Integrate? For Effective Ontology Merging Using K-means Clustering

Contact Info

Product

Resources

About