A proposal for improving spoken dialog systems using context information fusion

Context information fusion allow developing systems that provides large quantities of information obtained from network sensors with heterogeneous characteristics, that needs to be efficiently processed. During the last years, this field has become increasingly relevant due to the growing importance and rapid advances in sensor technology that provide information from the environment. Many works has incorporate context into a fusion data framework, showing the improvement of accuracy by integrating context, others proved the usefulness of context to improve classification task by using multiple classifier system. This paper proposes an improvement of a spoken dialog system using context information fusion. An effective application of this approach is described for a spoken dialog system providing railway information.