About https://w3id.org/scholarlydata/inproceedings/eswc2016/paper/research/40

Subject: https://w3id.org/scholarlydata/inproceedings/eswc2016/paper/research/40Property: http://www.w3.org/1999/02/22-rdf-syntax-ns#typehttp://www.w3.org/2000/01/rdf-schema#Resourcehttp://www.w3.org/2002/07/owl#Thinghttp://www.ontologydesignpatterns.org/ont/dul/DUL.owl#InformationObjecthttp://www.ontologydesignpatterns.org/ont/dul/DUL.owl#Objecthttps://w3id.org/scholarlydata/ontology/conference-ontology.owl#InProceedingshttp://www.ontologydesignpatterns.org/ont/dul/DUL.owl#SocialObjecthttp://purl.org/spar/fabio/ProceedingsPaperProperty: http://www.w3.org/2000/01/rdf-schema#labelDataset Recommendation for Data Linking: an Intensional ApproachProperty: http://www.w3.org/2002/07/owl#sameAshttp://data.semanticweb.org/conference/eswc/2016/paper/research/40Property: http://swrc.ontoware.org/ontology#abstractWith the increasing quantity and diversity of publicly available Web datasets, most notably, Linked Open Data, recommending datasets, which meet specific criteria, has become an increasingly important, yet challenging problem. Dataset recommendation is an important task when addressing issues such as entity retrieval, semantic search and, particularly, data linking, where one aims to identify datasets, which are likely to contain linking candidates. While an understanding of the nature of the content of specific datasets is a crucial prerequisite, we adopt the notion of dataset profiles, where a dataset is characterized through a set of schema concept labels that best describe it. Alternatively, by retrieving the textual descriptions of each of these labels, we can map the profiles to text documents. We introduce a dataset recommendation approach to identify linking candidates. The method is based on the presence of schema overlap between datasets by computing a semantico-frequential similarity between profiles and a ranking criterium based on the tf-idf cosine similarity. Our experiments, applied to all available Linked Datasets in the Linked Open Data (LOD) cloud, show that our method achieve an average precision up to 53% for a recall of 100%. An additional contribution of this work is the mapping returned for the schema concepts between datasets which are particularly useful in the linking step.Property: http://purl.org/dc/elements/1.1/creatorhttps://w3id.org/scholarlydata/person/konstantin-todorovhttps://w3id.org/scholarlydata/person/mohamed-ben-ellefihttps://w3id.org/scholarlydata/person/zohra-bellahsenehttps://w3id.org/scholarlydata/person/stefan-dietzeProperty: http://purl.org/dc/elements/1.1/titleDataset Recommendation for Data Linking: an Intensional ApproachProperty: http://purl.org/ontology/bibo/authorListhttps://w3id.org/scholarlydata/inproceedings/eswc2016/paper/research/40/authorListProperty: http://xmlns.com/foaf/0.1/makerhttps://w3id.org/scholarlydata/person/stefan-dietzehttps://w3id.org/scholarlydata/person/mohamed-ben-ellefihttps://w3id.org/scholarlydata/person/konstantin-todorovhttps://w3id.org/scholarlydata/person/zohra-bellahseneProperty: https://w3id.org/scholarlydata/ontology/conference-ontology.owl#abstractWith the increasing quantity and diversity of publicly available Web datasets, most notably, Linked Open Data, recommending datasets, which meet specific criteria, has become an increasingly important, yet challenging problem. Dataset recommendation is an important task when addressing issues such as entity retrieval, semantic search and, particularly, data linking, where one aims to identify datasets, which are likely to contain linking candidates. While an understanding of the nature of the content of specific datasets is a crucial prerequisite, we adopt the notion of dataset profiles, where a dataset is characterized through a set of schema concept labels that best describe it. Alternatively, by retrieving the textual descriptions of each of these labels, we can map the profiles to text documents. We introduce a dataset recommendation approach to identify linking candidates. The method is based on the presence of schema overlap between datasets by computing a semantico-frequential similarity between profiles and a ranking criterium based on the tf-idf cosine similarity. Our experiments, applied to all available Linked Datasets in the Linked Open Data (LOD) cloud, show that our method achieve an average precision up to 53% for a recall of 100%. An additional contribution of this work is the mapping returned for the schema concepts between datasets which are particularly useful in the linking step.Property: https://w3id.org/scholarlydata/ontology/conference-ontology.owl#hasAuthorListhttps://w3id.org/scholarlydata/inproceedings/eswc2016/paper/research/40/authorListProperty: https://w3id.org/scholarlydata/ontology/conference-ontology.owl#isPartOfhttps://w3id.org/scholarlydata/conference/eswc/2016/proceedingsProperty: https://w3id.org/scholarlydata/ontology/conference-ontology.owl#titleDataset Recommendation for Data Linking: an Intensional Approach
Subject: https://w3id.org/scholarlydata/person/zohra-bellahseneProperty: http://xmlns.com/foaf/0.1/madehttps://w3id.org/scholarlydata/inproceedings/eswc2016/paper/research/40
Subject: https://w3id.org/scholarlydata/person/stefan-dietzeProperty: http://xmlns.com/foaf/0.1/madehttps://w3id.org/scholarlydata/inproceedings/eswc2016/paper/research/40
Subject: https://w3id.org/scholarlydata/person/konstantin-todorovProperty: http://xmlns.com/foaf/0.1/madehttps://w3id.org/scholarlydata/inproceedings/eswc2016/paper/research/40
Subject: https://w3id.org/scholarlydata/person/mohamed-ben-ellefiProperty: http://xmlns.com/foaf/0.1/madehttps://w3id.org/scholarlydata/inproceedings/eswc2016/paper/research/40
Subject: https://w3id.org/scholarlydata/conference/eswc/2016/proceedingsProperty: https://w3id.org/scholarlydata/ontology/conference-ontology.owl#hasParthttps://w3id.org/scholarlydata/inproceedings/eswc2016/paper/research/40