PrSH: A belief description logic

Abstract

Some research has been done on probabilistic extension of description logics such as P-CLASSIC and P-SHOQ which focus on the statistical information. For example, in those kind of probabilistic DL, we can express such kind of uncertainty that the probability a randomly chosen individual in concept C is also in concept D is 90 percent. This kind of statistical knowledge is certain which means the author of this statement is sure about it. In this paper, we will describe a new kind of probabilistic description logic PrSH which could let user express the uncertain knowledge(i.e. degrees of belief). For example, if the user is not sure about that concept C is subsumed by concept D, he could describe it with PrSH such as the probability that concept C is subsumed by concept D is 90 percent.Furthermore, user could make use of the uncertain knowledge to infer some implicit knowledge by the extension of tableau-algorithm of SH which will be also introduced in this paper. © Springer-Verlag Berlin Heidelberg 2007.