Author(s): Alan Meehan, Kaniz Fatema, Eamonn Clinton, Lorraine McNerney, Declan O’Sullivan
Abstract: As Linked Data technologies mature and their advantages become more noteworthy, institutions are seeking to exploit these advantages and serve their closed (commercially valuable) data to customers on the web. To ensure customers are not accessing unauthorized data, an access control solution is required. In this paper we present a fine grain and flexibly access control approach for geospatial Linked Data. Our approach provides vocabularies to model licenses and templates. A license models what a customer is allowed to access at a fine granularity. A template models how data can be accessed. Each template contains variables and the values that these variables are allowed to contain are specified. This provides great flexibility and allows the creation of templates for a range of different data access use cases. Our approach is a web service which sits on top of a triple store and SPARQL processor and provides a RESTful API for access. RESTful calls to the service comprise of a license, a template and template variable values. Access control takes place by ensuring that the details specified in the RESTful call are allowed according to the license specified in the call. The paper details all the aspects of our access control approach. We also present a case study where a prototype implementation of our access control approach was applied to Ordnance Survey Ireland’s closed geospatial Linked Data.
Keywords: Linked Data; Access Control; Geospatial; Ordnance Survey Ireland