Ontology-based question answering for digital libraries

Abstract

In this paper we present an approach to question answering over heterogeneous knowledge sources that makes use of different ontology management components within die scenario of a digital library application. We present a principled framework for integrating structured metadata and unstructured resource content in a seamless manner which can then be flexibly queried using structured queries expressed in natural language. The novelty of the approach lies in the combination of different semantic technologies providing a clear benefit for the application scenario considered. The resulting system is implemented as part of the digital library of British Telecommunications (BT). The original contribution of our paper lies in the architecture we present allowing for the non-straightforward integration of the different components we consider. © Springer-Verlag Berlin Heidelberg 2007.