ADMI: a multi-agent architecture to autonomously generate data mining services

This paper presents a case for an intelligent agent based framework for knowledge discovery in a distributed environment comprising multiple heterogeneous data repositories. Data-mediated knowledge discovery, especially from multiple heterogeneous data resources, is a tedious process and imposes significant operational constraints on end-users. We demonstrate that autonomous, reactive and proactive intelligent agents provide an opportunity to generate end-user oriented, packaged, value-added decision-support/strategic planning services for professionals, managers and policy makers of an organization, without the need for a priori technical knowledge. Since effective progress of an organization is grounded in good communication, experience sharing, continuous learning and proactive actions, we present an agent-based data mining info-structure (ADMI) that deploys a suite of data mining (DM) algorithms coupled with intelligent agents to facilitate data access, DM query specification, DM algorithm selection and DM result visualization-i.e. automated generation of data-mediated decision-support/strategic-planning services.