Matches in ScholarlyData for { <https://w3id.org/scholarlydata/inproceedings/iswc2008/paper/research/328> ?p ?o. }
Showing items 1 to 16 of
16
with 100 items per page.
- 328 creator claudia-damato.
- 328 creator floriana-esposito.
- 328 creator nicola-fanizzi.
- 328 type InProceedings.
- 328 label "Statistical Learning for Inductive Query Answering on OWL Ontologies".
- 328 sameAs 328.
- 328 abstract "A family of parametric language-independent kernel functions defined for individuals in OWL ontologies is presented. These functions enable the exploitation of statistical learning for various ontology mining tasks. Namely, the novel kernel functions are integrated with a support vector machine for inducing simple mathematical models that can used for performing classification through an alternative mode w.r.t. logical reasoning. Besides we also present a method for adapting the kernel to the knowledge base through stochastic optimization. An inductive approach based on machine learning may be useful especially to cope with the inherent incompleteness of the knowledge bases. The system implementing the whole framework has been tested in some experiments on query answering with real ontologies drawn from standard repositories.".
- 328 hasAuthorList authorList.
- 328 hasTopic Semantic_Web.
- 328 isPartOf proceedings.
- 328 keyword "Approximate classification".
- 328 keyword "kernel methods".
- 328 keyword "knowledge discovery".
- 328 keyword "ontology mining".
- 328 keyword "statistical learning".
- 328 title "Statistical Learning for Inductive Query Answering on OWL Ontologies".