Extraction method of the mutual understanding gap based on egocentrism in short dialogues

Abstract

Question and answering sites have a lot of meaningless dialogues which are different from the intention of questioners, because many speakers speak in the same site at the same time. We tried to extract an egocentrism of the human from speeches of question and answering sites and showed the mutual understanding gap based on the method to solve this issue. In this paper, we propose the method to extract the mutual understanding gap in short dialogues such as two or less speeches that is not focused so far. As a result, the mutual understanding gap in wide-ranging dialogue data can be detected, and it is extensible to a further applications. © 2011 Springer-Verlag.