In the man-machine interfaces, it is important to use dialogue understanding technologies. One of the practical application fields is a question and answering (QA) systems. In order to reply appropriate answers for user's questions, this paper presents a dialogue technique by transforming semantic expressions for both requests and answers. The measurements for the disrepute of the QA system are introduced for requests and answers, respectively. For the KAMOKUMA QA system generating answers which are reflecting user's intension, the presented scheme is applied. For the AQ data with 7,518 requests, the real time simulation to estimate user's sufficiency is computed. © 2009 Springer Berlin Heidelberg.
CITATION STYLE
Inada, Y., Nakano, H., Kashiji, S., & Aoe, J. (2009). Intelligent QA systems using semantic expressions. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 5712 LNAI, pp. 308–316). https://doi.org/10.1007/978-3-642-04592-9_39
Mendeley helps you to discover research relevant for your work.