Abstract
Queryable Linked Data is published through several interfaces, including sparql endpoints and Linked Data documents. In October 2014, the dbpedia Association announced an official Triple Pattern Fragments interface to its popular dbpedia dataset. This interface proposes to improve the availability of live queryable data by dividing query execution between clients and servers. In this paper, we present a usage analysis between November 2014 and July 2015. In 9 months time, the interface had an average availability of 99.99 %, handling 16,776,170 requests, 43.0% of which were served from cache. These numbers provide promising evidence that low-cost Triple Pattern Fragments interfaces provide a viable strategy for live applications on top of public, queryable datasets.
Author supplied keywords
Cite
CITATION STYLE
Verborgh, R. (2015). DBpedia’s triple pattern fragments:Usage patterns and insights. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 9341, pp. 431–442). Springer Verlag. https://doi.org/10.1007/978-3-319-25639-9_54
Register to see more suggestions
Mendeley helps you to discover research relevant for your work.
