Exploiting Mathematical Structures of Statistical Measures for Comparison of RDF Data Cubes

A growing number of public institutions all over the world has recently started to publish statistical data according to the RDF Data Cube vocabulary, as open and machine-readable Linked Data. Although this approach allows easier data access and consumption, appropriate mechanisms are still needed to perform proper comparisons of statistical data. Indeed, the lack of an explicit representation of how statistical measures are calculated still hinders their interpretation and use. In this work, we discuss an approach for the analysis and schema-level comparison of distributed data cubes, which is based on the formal and mathematical representation of measures. Relying on a knowledge model, we present and evaluate a set of logic-based functionalities able to support novel typologies of comparison of different data cubes.