Generating tailored textual summaries from ontologies

Abstract

This paper presents the ONTOSUM system which uses Natural Language Generation (NLG) techniques to produce textual summaries from Semantic Web ontologies. The main contribution of this work is in showing how existing NLG tools can be adapted to Semantic Web ontologies, in a way which minimises the customisation effort while offering more diverse output than template-based ontology verbalisers. A novel dimension of this work is the focus on tailoring the summary formatting and length according to a device profile (e.g., mobile phone, Web browser). Another innovative idea is the use of ontology mapping for summary generation from different ontologies. © Springer-Verlag Berlin Heidelberg 2005.