Cargando…

Semantic search for novel information /

Detalles Bibliográficos
Clasificación:Libro Electrónico
Autor principal: Färber, Michael (Autor)
Formato: Electrónico eBook
Idioma:Inglés
Publicado: Amsterdam, Netherlands : IOS Press, [2017]
Colección:Studies on the Semantic Web ; v. 031.
Temas:
Acceso en línea:Texto completo
Tabla de Contenidos:
  • Title Page ; Abstract; Acknowledgements; Contents; List of Figures; List of Tables; List of Listings; Introduction; Motivation; Problem Statement; Research Questions; Contribution of the Thesis; Published Results; Readers' Guide; Foundations; Semantic Web Technologies; The Vision of the Semantic Web; RDF and SPARQL; Knowledge Graph; Information Extraction, Machine Learning, Information Retrieval, and Data Quality; Information Extraction; Machine Learning; Information Retrieval; Data Quality; State-of-the-Art; Statistical Search for Relevant Information; Temporal Information Retrieval.
  • Trend Detection; Semantic Search for Relevant Information; Semantic Search for Relevant Entities; Semantic Search for Relevant Statements; Semantic Search for Relevant Events; Statistical Search for Relevant, Novel Information; Characteristics of Statistical Search for Relevant, Novel Information; Evaluations and Data Sets; Approaches to the Statistical Search for Relevant, Novel Information; Semantic Search for Relevant, Novel Information; Semantic Search for Novel Entities; Semantic Search for Novel Statements; Semantic Search for Novel Events.
  • The Suitability of Knowledge Graphs for Semantic Novelty Detection; Selection of Knowledge Graphs; Key Statistics of Selected Knowledge Graphs; Related Work; Number of Triples and Statements; Classes and Domains; Relations and Predicates; Instances and Entities; Subjects and Objects; Summary of Key Statistics; Completeness and Timeliness of Selected Knowledge Graphs; Gold Standard; Completeness; Timeliness; Discussion; Conclusions; Emerging Entity Detection; Motivation; Entity Linking Challenges Arising from Missing Entities and Missing Surface Forms; Overview of Entity Linking Challenges.
  • Challenges in the Wild; Summary of Findings; Approach: Emerging Entity Detection; The Approach; Evaluation Results; Related Work; Challenge 1: Linking to in-KG Entities via Known Surface Forms; Challenge 2: Linking to in-KG Entities via Unknown Surface Forms; Challenge 3: Linking to Out-of-KG Entities via Known Surface Forms; Challenge 4: Linking to Out-of-KG Entities via Unkown Surface Forms; Conclusions; Novel Statement Extraction; Motivation; Measuring Semantic Novelty of Statements; The Novel Statement Extraction System; Textual Triple Extraction; KG Linking; Novelty Detection.
  • Evaluation 1: CrunchBase; Data Used; Evaluation Setting; Evaluation Results; Evaluation 2: DBpedia; Data Used; The Baseline Approach and its Evaluation Results; Evaluation Results of Our Approach; Discussion; Related Work; Conclusions; Conclusions; Summary; Limitations; Outlook; Appendix; Supplementary Material; Emerging Entity Detection; Bibliography.