Knowledge management system: an agent-based approach

Knowledge is an important asset in any enterprise because of global competition and the rapid development of information technology. Knowledge management (KM) is viewed as an important factor in improving the competitive edge of an enterprise. By its very nature, knowledge is disparate and heterogeneous and can be represented in various ways (text, pdf, html, etc.), and can be either structured or unstructured. It is, therefore, difficult to acquire, organize or distribute knowledge using only traditional information technology methods such as e-mail or file servers. Because of the autonomous and collaborative aspects inherent in agent-based technology, this may be a possible solution to the problem. In this paper, an agent-based system is proposed to conceptualize the activities of KM and an annotation process is developed to address the heterogeneity issue of knowledge sources. Moreover, an agent conversation policy, which makes agent communication more effective, is proposed. This agent-based system shows great promise in KM and the conversation policy enhances communication between agents in a heterogeneous environment.