Agent-Mediated Knowledge Management
Papers from the AAAI Spring Symposium
Ludger van Elst, Virginia Dignum, and Andreas Abecker, Cochairs
Knowledge management (KM) has been a predominant trend in business in recent years. Not only is Knowledge Management an important field of application for AI and related techniques, such as CBR technology for intelligent lessons-learned systems, it also provides new challenges to the AI community, like context-aware knowledge delivery. Scaling-up research prototypes to real-world solutions usually requires an application-driven integration of several basic technologies, e.g., ontologies for knowledge sharing and reuse, collaboration support, and personalized information services. Typical characteristics to be dealt with include manifold logically and physically dispersed actors and knowledge sources; different degrees of formalization of knowledge; different kinds of (web-based) services and (legacy) systems, and conflicts between local (individual) and global (group or organizational) goals.
Agent approaches have already been successfully employed for many partial solutions within the overall picture: Agent-based workflow, cooperative information gathering, intelligent information integration, or personal information agents are established techniques in this area. In order to cope with the inherent complexity of a more comprehensive solution, the concept of Agent-Mediated Knowledge Management (AMKM) deals with collective aspects of the domain in an attempt to cope with the conflict between desired order and actual behavior in dynamic environments. AMKM introduces a social layer, which structures the society of agents by defining specific roles and possible interactions between them. In this symposium contributions illustrate methodological, technical and application aspects of AMKM.