Towards Odalic, a Semantic Table Interpretation Tool in the ADEQUATe Project

The goal of the ADEQUATe project is to assess and improve quality of the (tabular) open data being published at two Austrian open data portals – https://www.data.gv.at and https://www.opendataportal.at. The goal of the quality improvement technique described in this paper is to semantically interpret such tabular data and publish them as Linked Data; this basically means to (1) classify columns of the input data using Linked Data vocabularies, (2) link cell values of the input data against Linked Data entities, (3) discover relations among the columns of the input data by searching for evidences of such relations among Linked Data sources, and (4) export such semantically interpreted data as RDF/Linked Data. In this paper, we describe limitations of TableMiner+, one of the tools for semantic table interpretation, with respect to our needs in the ADEQUATe project. Furthermore, we present Odalic, a tool for (semi)automatic semantic table interpretation and Linked Data publishing and describe how it addresses these limitation. We describe lessons learned using Odalic in the ADEQUATe project and also future work planned.