Using the UML to Model Knowledge in Agent Systems

摘要

The UML has been used to model the knowledge required to engineer agent systems. The UML provides mechanisms for modelling the knowledge about a domain, partitioning the knowledge into appropriate packages and providing a means of moving from analysis to design. This paper argues that the UML is a suitable language for knowledge modelling in agent systems and provides details of the techniques and extensions employed. By modifying and applying commonly used mainstream modelling languages such as the UML tool and methodology support for the construction of agent systems can be leveraged from existing practices.