Abstract:
The Resource Description Framework (RDF) is a general framework for the standardised interchange of information over the semantic web by way of RDF triples. Specifying temporal information which describes when individual triples hold using standard RDF is convoluted, and temporal query evaluation is inefficient. While there has been work on developing temporal triple stores that address these issues, there is generally a large cost associated with migrating to a new triplestore, especially if that triplestore is not being actively maintained. In this PhD we propose a query language named LSPARQL and a query evaluation engine that partially sidesteps this issue by executing queries over a standard transaction log which are commonly supported by most triplestores. We provide a formal definition of the semantics of a log-based data model and of our query language. We describe a proof-of-concept implementation that demonstrates that the average triplestore could support transaction time temporal queries relatively easily without having to redesign their data model.