Current approaches to the interoperability of heterogeneous resources typically maintain coordinated clones over which querying infrastructure operates. In the case of large-scale repositories, and especially when no single schema is clearly established so that the problem can be reduced to transforming legacy data, a more dynamic approach would be a benefit. In many real-world situation, how- ever, useful queries need to combine information from different sources that are actively maintained in incompatible schemata and are too large to systematically clone. Examples include the various agricultural resource repositories and databases such as meteorological archives and GIS.
In this position paper we explore the applicability of the W3C Protocol for Web Description Resources (POWDER) as infrastructure for the efficient and distributed retrieval of the meta-information needed to dynamically re-write queries in one schema to the (set of) semantically equivalent queries that need to be executed over the various heterogeneous schemata. This enables information providers to publish information in any schema as long as POWDER is used to annotate their repository with the coordination-related meta-information. Querying engines can then exploit existing and established resource discovery mechanisms implemented over the POWDER protocol to retrieve the meta-information pertinent to any single triple pattern in a query and use that to dynamically perform query rewriting.
Contributed by Pythagoras Karampiperis, Stasinos Konstantopoulos and Vangelis Karkaletsis.
View/download submitted paper from HERE.
Click HERE to evaluate the paper.
In this position paper we explore the applicability of the W3C Protocol for Web Description Resources (POWDER) as infrastructure for the efficient and distributed retrieval of the meta-information needed to dynamically re-write queries in one schema to the (set of) semantically equivalent queries that need to be executed over the various heterogeneous schemata. This enables information providers to publish information in any schema as long as POWDER is used to annotate their repository with the coordination-related meta-information. Querying engines can then exploit existing and established resource discovery mechanisms implemented over the POWDER protocol to retrieve the meta-information pertinent to any single triple pattern in a query and use that to dynamically perform query rewriting.
Contributed by Pythagoras Karampiperis, Stasinos Konstantopoulos and Vangelis Karkaletsis.
View/download submitted paper from HERE.
Click HERE to evaluate the paper.
A note to the authors: Section 3.3 needs apparently a (re-)write! It only contains one paragraph, which is actually a repetition of the 1st paragraph of the previous section...
ReplyDeleteThis is a pity, since this section would be probably the most interesting piece of the paper.