Philosophy and methodology for knowledge discovery in autonomic computing systems

Autonomic computing has been advanced as a solution to the currently problematic control of voice and data communications networks. Autonomic systems adapt both to their environment and to the demands placed upon them as a consequence of the use of the system(s) within their purview. Data and voice networks function in a changing environment with varying use cases; hence, autonomic systems must be deployed with both a significant a priori knowledge base and the capability to continuously upgrade that knowledge base. The system must engage in some amount of unsupervised learning and hypothesize as to nature of its functioning. Maintenance of hypotheses and theories is intrinsic to the system, especially in evolutionary scenarios. This paper explores how knowledge maintenance is done for voice and data communications networks applications that use autonomic system approaches.