Authors: Dmitri Roussinov, Jose Antonio Robles-Flores
Addresses: Department of Information Systems, W.P. Carey School of Business, Arizona State University, P.O. Box 874606, Tempe, AZ 85287–4606, USA. ' ESAN University, Lima, Peru and Department of Information Systems, W.P. Carey School of Business, Arizona State University, P.O. Box 874606, Tempe, AZ 85287–4606, USA
Abstract: We introduce a novel and completely trainable approach to automated open domain Question Answering (QA) on the web, for the purpose of business intelligence. Our approach does not involve any linguistic resources and can be easily implemented within any information awareness system. When tested on standard test collections, the performance of our approach was found to be comparable with that of the current top-of-the-line, complex and expensive linguistic approaches, and significantly better than that of other completely trainable approaches. We also present the design of our ongoing empirical study and the qualitative observations from our pilot experiments.
Keywords: information retrieval; open domain question answering; web search; business intelligence; natural language questions.
International Journal of Internet and Enterprise Management, 2005 Vol.3 No.1, pp.46 - 62
Published online: 25 Jun 2005 *Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article