Publication and Usage of Official Czech Pension Statistics Linked Open Data

Abstract

Linked Open Data (LOD) principles are known for a decade now and there are thousands of LOD datasets of variable quality and importance. They often come from academic research projects, which show how LOD can be published, how it can be useful, etc. However, these projects last only for a few years and when they end, the datasets often cease to be maintained. What is needed is to convince the owner of the original data, e.g. an organization in public administration, to keep on publishing LOD on their own even when the project ends. Therefore, it is noteworthy and admirable, when this happens. In this paper we describe how the Czech Social Security Administration (CSSA) publishes official pension statistics as LOD as part of their day-to-day operation, which was jump-started by an applied research project. The data is modeled using the Simple Knowledge Organization System (SKOS) vocabulary and the RDF Data Cube Vocabulary (DCV). It is published as RDF data dumps, as a SPARQL endpoint and using IRI dereferencing for semantic web technologies power users. For journalists and other users without knowledge of these technologies, the data is also published as CSV files and visualizations generated from the LOD. We show how the data is reused in applications and how it contributes to statistical indicators in combination with other LOD.

Keywords: Pension, Statistics, Data Cube Vocabulary, Linked Data