From the OntoBayes Model to a Service Oriented Decision Support System Yi Yang and Jacques Calmet Institute for Algorithms and Cognitive Systems (IAKS) University of Karlsruhe (TH) 76131 Karlsruhe, Germany {yiyang,calmet}@ira.uka.de Abstract The aim of this paper is to propose a service oriented decision support system based on an ontology-driven un- certainty model (OntoBayes). OntoBayes consists of knowl- edge and decision model parts. The former is the integra- tion of ontologies and Bayesian Networks while the latter can describe different decision models. OntoBayes gives a solution to deal with uncertainty and structure complexity and provides decision models by decision making. In order to construct a decision support system, a service oriented framework and architecture will be introduced finally. 1. Introduction During the 1950s and 1960s the concept of decision sup- port was investigated from two aspects: the theory of orga- nizational decision making and the techniques of interactive computer systems. After that it became an area of research that focuses on computerized system supporting decision making activities [1]. A decision support system (DSS) can be defined as “a computer program that provides informa- tion in a given domain of application by means of analytical decision models ...” [2]. Nowadays there are different types of DSS at the conceptual level: communication-driven DSS, data-driven DSS, document-driven DSS, knowledge-driven DSS, and model-driven DSS [3]. In this paper, we propose a service oriented DSS. First, we investigate an ontology driven uncertain model, Onto- Bayes, consisting of knowledge and decision model parts. The former is the integration of ontologie and Bayesian Net- works (BN) while the latter can describe different decision models. After that a service oriented design of DSS at a high level is introduced. The selected overall framework for this DSS is the multiagent paradigm. The remaining sections of this paper are structured as follows. Section 2 scratches the surface of the agent ap- proach for web services and decision making. Section 3 introduces an ontology driven uncertain model OntoBayes. This section differs partly from previously published results since it is now fully suited to decision support. Section 4 is devoted to investigate the design of a service oriented de- cision support system based on OntoBayes. Section 5 sur- veys related works and finally, Section 6 contains a brief overview on future work and concludes the paper. 2. Agent, Web Services and Decision Making In the cyber world, agent based approaches are more powerful when running agents in a distributed and dynamic environment (potentially on a web-wide scale) to perform complex actions on behalf of their users [4]. Unifying agents and web services can enhance the construction and flexibility of web service applications [5]. Before introducing other agent based approaches in the following sections, it is necessary to give some agents re- lated definitions hereafter. An agent is capable of au- tonomous action in situated environment in order to meet its design objectives [6]. Based on this definition an intel- ligent agent can be extended with three additional charac- teristics: reactivity, proactivity and social ability. The con- cept of multiagent has emerged as a paradigm for designing complex software systems. It is used to better formalize problems in Distributed Artificial Intelligence (DAI) [7]. The world of web services was characterized as loosely- coupled distributed systems based on service oriented com- puting (SOC). The use of web services could be consid- ered as actions that the agent may take to meet its goals. In [8] four major trends in internet computing were analyzed which are able to driven SOC and Multi Agent Systems (MAS) research in the future. There are some emerging approaches with MAS-like characteristics in SOC, such as ubiquitous computing, ontologies, service-level agreements and quality-of-service measures etc. All of them can be per- fectly performed with MAS concepts and techniques. In general, decision making is not a simple event but a process leading to the selection of a course of action among International Conference on Computational Intelligence for Modelling Control and Automation,and International Conference on Intelligent Agents,Web Technologies and Internet Commerce (CIMCA-IAWTIC'06) 0-7695-2731-0/06 $20.00 © 2006