Ontology Engineering Based on Spreadsheet Data Transformation

Abstract

Spreadsheet data are an important source of information and can be used to develop domain ontologies. In this paper, we describe a novel approach for ontology engineering based on spreadsheet data transformation. A feature of the proposed approach is the automatic synthesis of a complete ontology in the OWL format both at the level of an ontological schema and at the level of instances (assertion). Our approach consists of five main stages: converting source spreadsheet to JSON representation and data cleaning; defining column types and a subject column; generating OWL ontology fragments; aggregating received OWL ontology fragments. The approach is implemented in the form a command-line tool. We also present an experimental evaluation of the implemented tool for generating ontology by transforming test spreadsheet data. The dataset of test spreadsheets called Tough Tables (2T) is used for our experiment. We obtained high scores of precision and recall for this dataset, which shows the applicability of our solution for automatic generation of domain ontologies.