Modeling uncertainties in advanced knowledge management

Abstract

Handling and managing knowledge is a difficult and complex enterprise. a wide range of advanced technologies have to be invoked in providing assistance for knowledge requirements ranging from acquisition, modeling, (re)using, retrieving, publishing and maintaining of knowledge. Knowledge engineers swap ideas, communicate, plan, act, or reason often in situations where facts are unknown and the underlying natural language is uncertain or vague. They do not have access to the complete environment to evaluate each situation. Also conditions are unknown, incomplete or only crudely summarized. In this paper, knowledge representation applications as formal graphical language are examined in detail. exempli gratia, bayesian networks are graphical models to represent knowledge under conditions of uncertainty. This network type model the quantitative strength of the connections between variables allowing probabilistic beliefs about them to be updated automatically as new information becomes available. Applications in various fields like mechanical engineering are exemplified. © 2011 Springer-Verlag.