HTML5 based Facet Browser for SPARQL Endpoints

Abstract

The Linked Data concept uses a collection of Semantic Web technologies in order to interconnect, publish, and share pieces of data on the Web, in a machine-readable format. It enables querying and combining data from different datasets together in order to retrieve specific information and enable usecase scenarios which are unavailable over isolated datasets. However, the process of querying linked data published on different places on the Web poses several challenges. Generally, each user should know the schema of the data, write a query and access a relevant linked data endpoint. According to statistics, endpoints are often unavailable and have significant downtime, so this presents a serious obstacle in application development and scenarios which rely on the data. In this paper, we present a facet browser for SPARQL endpoints, based on HTML5. It allows users to search and retrieve RDF triples based on a keyword, from public SPARQL endpoints. By using HTML5 Web Storage, the triples from the results can be saved in the browser, locally, for future use. The facet browser provides management functionalities over the stored data - capabilities to update, refresh, modify, delete and download the triples in various RDF formats: JSON-LD, Turtle, NTriples, RDF/XML, JSON, CSV. The locally stored RDF triples can also be shared with other users. We believe that these features of the facet browser will help overcome the endpoint downtime issues, by providing offline data accessibility for the user and his applications.