Detecting and Reporting Extensional Concept Drift in Statistical Linked Data

Abstract

The RDF Data Cube vocabulary is a catalyst for the availability of statistical Linked Data: raw statistical Linked Data are easy to model in, publish to, and retrieve from the Linked Data cloud. In statistical datasets, concepts are central entities represented by variables and their values. The meaning of these concepts is often assumed to be stable, but in fact it can change over time: we call this concept drift. Extensional concept drift is one type of change of meaning that affects the things the concept extends to. It occurs frequently in historical datasets, and it can have drastic consequences on longitudinal querying. In this paper we propose and use a method to detect extensional concept drift in a dataset modelled using the RDF Data Cube vocabulary: the Dutch historical censuses. We analyze, model and publish back the occurrence of extensional concept drift in concepts of the occupation census, advocating straightforward publishing of results in a pull-push workflow.