Web Question Answering: Technology and Business Applications
Dmitri G. Roussinov, José Antonio Robles-Flores · Journal of the Association for Information Systems · 2004
While being successful in providing keyword based access to web pages, commercial search portals, such as Google, Yahoo, AltaVista, and AOL, still lack the ability to answer questions expressed in a natural language.We have explored the feasibility of a completely trainable approach to automated question answering on the Web for the purpose of business intelligence and other practical applications.We introduce an entirely self-learning approach that does not involve any linguistic resources.It can be easily implemented within various information awareness systems.The performance of our approach was found comparable to (and even relatively better than) many other more complex and expensive approaches.We also present the design of our empirical study and the qualitative observations from our pilot experiments.