KDPMEL: A Knowledge Discovery Process Modeling and Enacting Language

Abstract

Knowledge Discovery in Databases (KDD) is a complex process that is highly interactive and iterative. Managing the process and its underlying resources and interacting activities is a very complex task. Today, KDD projects are typically approached in an unstructured, ad hoc manner due to a lack of formal methods and methodologies for their development. The similarities between KDD processes and software development processes suggest that approaches used to manage the development of software processes, such as process programming and processcentered support environments, are also applicable and in fact advantageous to KDD processes. This paper proposes the Knowledge Discovery Process Modeling and Enacting Language (KDPMEL) that is designed specifically to provide both the syntax and semantics needed to capture and execute KDD processes. KDPMEL combines aspects of both process programming and KDD to represent KDD processes precisely as process programs. Along with KDPMEL, an Integrated Development Environment (IDE) for KDPMEL is proposed to enable and facilitate the development and execution of KDPMEL programs. We illustrate and evaluate KDPMEL by representing and executing illustrative examples of KDD activities and tasks