A Conceptual Infrastructure for Knowledge Discovery Process with Granularity

Knowledge discovery has received more and more attention from the business community for the last few years. One of the most important and challenging problems in it is the definition of discovery process model, which are well understood, efficiency, and quality of outcome. A conceptual infrastructure for knowledge discovery process is proposed with business understanding, model selection and domain knowledge integration in evolving database environment. The corresponding layered architecture creates an automatable process and communication mechanism to incorporate known knowledge and data granules into discovery process, through ontology service facility. A higher order mining method embedded in the process is proposed, to achieve monitoring and identifying changes statically and tracing trends dynamically. Finally, implementation techniques of basic data structure and simplified mining algorithms are discussed.