W3C PROV to describe provenance at the dataset, feature and attribute levels in a distributed environment. (July 2017)

• Record Type: Journal Article
• Authors: Closa, Guillem; Masó, Joan; Proß, Benjamin; Pons, Xavier
• Is Part Of: Computers, environment and urban systems. Volume 64(2017)
• Page Start: 103
• Page End: 117
• Publication Date: 2017-07
• Subjects: Provenance -- Distributed enviroment -- Conflation -- GIS -- Metadata; City planning -- Data processing -- Periodicals; Regional planning -- Data processing -- Periodicals; 303.4834
• DOI: 10.1016/j.compenvurbsys.2017.01.008
• Languages: English
• ISSNs: 0198-9715

Abstract

Abstract: Provenance, a metadata component referring to the origin and the processes undertaken to obtain a specific geographic digital feature or product, is crucial to evaluate the quality of spatial information and help in reproducing and replicating geospatial processes. However, the heterogeneity and complexity of the geospatial processes, which can potentially modify part or the complete content of datasets, make evident the necessity for describing geospatial provenance at dataset, feature and attribute levels. This paper presents the application of W3C PROV, which is a generic specification to express provenance records, for representing geospatial data provenance at these different levels. In particular, W3C PROV is applied to feature models, where geospatial phenomena are represented as individual features described with spatial (point, lines, polygons, etc.) and non-spatial (names, measures, etc.) attributes. This paper first analyses the potential for representing geospatial provenance in a distributed environment at the three levels of granularity using ISO 19115 and W3C PROV models. Next, an approach for applying the generic W3C PROV provenance model to the geospatial environment is presented. As a proof of concept, we provide an application of W3C PROV to describe geospatial provenance at the feature and attribute levels. The use case presented consists of a conflation of the U.S. Geological Survey dataset with the National Geospatial-Intelligence Agency dataset. Finally, an example of how to capture the provenance resulting from workflows and chain executions with PROV is also presented. The application uses a web processing service, which enables geospatial processing in a distributed system and allows to capture the provenance information based on the W3C PROV ontology at the feature and attribute levels. Highlights: Analyses the potential for representing geospatial provenance at the three levels of granularity using ISO 19115 and W3C PROV models. An approach for applying the generic W3C PROV provenance model to the geospatial environment is presented. Full ontology description for representing geospatial provenance with W3C PROV. The use case presented consists on a conflation of the U.S. Geological Survey dataset with the National Geospatial-Intelligence Agency dataset. Example of how to capture the provenance resulted from workflows and chain executions with PROV and it technological architecture is presented.