@prefix foaf: . @prefix ns1: . ns1:wolfgang-nejdl foaf:made . ns1:stefan-dietze foaf:made . @prefix ns2: . @prefix ns3: . ns3:proceedings ns2:hasPart . @prefix rdf: . rdf:type ns2:InProceedings . @prefix rdfs: . rdf:type rdfs:Resource . @prefix owl: . rdf:type owl:Thing . @prefix ns7: . rdf:type ns7:InformationObject , ns7:Object , ns7:SocialObject . @prefix ns8: . rdf:type ns8:ProceedingsPaper ; rdfs:label "A Scalable Approach for Efficiently Generating Structured Dataset Topic Profiles" ; owl:sameAs . @prefix ns9: . ns9:abstract "The increasing adoption of Linked Data principles has led to an abundance of datasets on the Web. However, take-up and reuse is hindered by the lack of descriptive information about the nature of the data, such as their topic coverage, dynamics or evolution. To address this issue, we propose an approach for creating linked dataset profiles. A profile consists of structured dataset metadata describing topics and their relevance. Profiles are generated through the configuration of techniques for resource sampling from datasets, topic extraction from knowledge bases and their ranking based on graphical models. To enable a good trade-off between scalability and representatives of generated data, appropriate parameters are determined experimentally. Our evaluation considers topic profiles of all accessible datasets from the Linked Open Data cloud and shows that our approach generates representative profiles even with comparably small sample sizes (10%) outperforms established topic modelling approaches." . @prefix dc: . dc:creator ns1:wolfgang-nejdl , ns1:davide-taibi , ns1:bernardo-pereira-nunes , ns1:besnik-fetahu , ns1:marco-antonio-casanova , ns1:stefan-dietze ; dc:subject "Linked Data" , "Metadata" , "Profiling" , "Vocabulary of Links" ; dc:title "A Scalable Approach for Efficiently Generating Structured Dataset Topic Profiles" . @prefix ns11: . @prefix ns12: . ns11:authorList ns12:authorList ; foaf:maker ns1:wolfgang-nejdl , ns1:stefan-dietze , ns1:bernardo-pereira-nunes , ns1:besnik-fetahu , ns1:marco-antonio-casanova , ns1:davide-taibi ; ns2:abstract "The increasing adoption of Linked Data principles has led to an abundance of datasets on the Web. However, take-up and reuse is hindered by the lack of descriptive information about the nature of the data, such as their topic coverage, dynamics or evolution. To address this issue, we propose an approach for creating linked dataset profiles. A profile consists of structured dataset metadata describing topics and their relevance. Profiles are generated through the configuration of techniques for resource sampling from datasets, topic extraction from knowledge bases and their ranking based on graphical models. To enable a good trade-off between scalability and representatives of generated data, appropriate parameters are determined experimentally. Our evaluation considers topic profiles of all accessible datasets from the Linked Open Data cloud and shows that our approach generates representative profiles even with comparably small sample sizes (10%) outperforms established topic modelling approaches." ; ns2:hasAuthorList ns12:authorList ; ns2:isPartOf ns3:proceedings . @prefix xsd: . ns2:keyword "Vocabulary of Links"^^xsd:string , "Linked Data"^^xsd:string , "Profiling"^^xsd:string , "Metadata"^^xsd:string ; ns2:title "A Scalable Approach for Efficiently Generating Structured Dataset Topic Profiles" . ns1:bernardo-pereira-nunes foaf:made . ns1:besnik-fetahu foaf:made . ns1:davide-taibi foaf:made . ns1:marco-antonio-casanova foaf:made .