| 364 Semiometrics: Applying Ontologies across Large-Scale Digital Libraries (2008) | |||||||||||||||
Abstract | |||||||||||||||
| Abstract. As large-scale digital libraries become more available and complete, not to mention more numerous, it is clear there is a need for services that can draw together and perform inference calculations on the metadata produced. However, the traditional Relational Database Management System (RDBMS) model, while efficiently constructed and optimised for many business structures, does not necessarily cope well with issues of concurrent data updates and retrieval at the scale of hundreds of thousands of papers. At the same time the growth of RDF and the increasing interest in Semantic Web technologies perhaps begins to present a viable alternative at a scalable, practical level. This paper considers a specific application of large-scale metadata analysis and conducts scalability tests using real-world data. It concludes that RDF technologies are both a scalable and performance-realistic alternative to traditional RDBMS approaches. It also shows that for relationship-based queries on large-scale metadata stores, RDF technologies can significantly out-perform traditional RDBMS approaches by allowing both retrieval and updating of data in a timely manner. 1 | |||||||||||||||
Publication details | |||||||||||||||
| |||||||||||||||