Engineering of Multi-Agent Systems to Effectuate Distributed Data Mining Activities Syed Zahid Hassan ZAIDI a , Syed Sibte Raza ABIDI b & Zafar Iqbal Hashmi c a,c Health Informatics Research Group, School of Computer Sciences Universiti Sains Malaysia, Penang, MALAYSIA b Faculty of Computer Science, Dalhousie University, Halifax B3H 1W5, CANADA Abstract The proliferation of healthcare data has resulted in a large number of concerted efforts to inductively discover ‘useful’ knowledge from the collected data, and indeed interesting results have been reported by health informatics researchers. We argue that with the existence of multiple heterogeneous data repositories in a healthcare enterprise we need to establish a distributed data community, such that any DM effort draws upon the ‘holistic’ data available within the entire healthcare enterprise. When adopting this view, a set of data access and mining issues can be addressed using the well-known software agent technology. The aim of this paper is to propose the methodology of multi-agents system and engineering of individual agent to effectuate distributed DM activities applied to heterogeneous healthcare repositories. Keywords: Intelligent agents technology; Multi-agents organization; Agents’ planning; and KQML. 1. Introduction To date, many knowledge discovery systems have been developed prior to emergence of agent-based computing. The main function of these systems are to statically mine multiple level of knowledge from relational databases [1], but most of systems are not able to provide the services, pertinent towards the improvement/progress of any organization. The operational efficacy of an organization can be significantly increased by acquiring empirical knowledge from data repositories and by operationalzing procured empirical knowledge to derive the suite of decision support services that aim to impact strategic decision-making, planning and management of the organization [7]. The vantage point of these services are that they provide insights to assists healthcare analyst and policies makers to make strategic decisions or predict future consequences by taking into account the actual outcomes of current operative values. Typically, the services may include: Analysis, planning, trending, examine, forecasting, predicting bench marking and best practices reporting, outcomes measurement, what–if scenario analysis, comparing organization practice with organization rules, market research , effectiveness on outcomes of treatment, data analysis for organization financing, health surveillance and resource allocations [7]. To meet the above objectives, deployment of new analytic functionalities and data access methods have now become necessary for the systems to become the full fledged members of the new information era. Manually changing the systems is a nontrivial task. A preferable way of system working is to construct agent wrappers around KDD systems [3]. These agent wrappers interface to the information sources and information consumer, providing a uniform way to accessing data as well as offering additional functionalities, such as monitoring the changes and provide the services on demand. The critical question then is how to structure and organize these multiple agents to achieve user centric goal.