Búsqueda avanzada

A Federated Approach for Array and Entity Environmental Linked Data


Available environmental and spatial data is increasing in size and every time new application domains take advantage of this fact. The need for accessing them through linked data paradigm is also increased, due to the interest of their combination with already available linked data repositories. Entity based environmental data fits perfectly to the graph data model of RDF, however, much environmental data are array-based, and such data are clearly not efficiently represented with RDF. In fact, transforming array environmental data to RDF triples in some datasets will generate huge RDF datasets. Querying these datasets through SPARQL will lead to low performance solutions. In this paper, we propose a federated architecture that integrates entity and array-based repositories into a single SPARQL-based framework, where SPARQL queries are translated into SQL and array-based queries. New operations will be added to SPARQL algebra in order to embed those relational and array-based queries into SPARQL query plans. This will make SPARQL able to access two different database paradigms (entity and array) in one query to answer questions like “What is the predicted average of temperature of each municipality of Spain for the next week?”

Palabras Clave:

Linked Data, Geospatial and environmental data, GeoSPARQL, SPARQL Query Processing





Este artículo tiene una licencia de uso CreativeCommons Reconocimiento (by)

Descarga el artículo haciendo click aquí.