Document Type



Knowledge discovery from large databases has become an emerging research topic and application area in recent years primarily because of the successful introduction of large business information systems to enterprises in the electronic business era. However, transferring subjects/problems from managerial perspective to data mining tasks from information technology perspective requires multidisciplinary domain knowledge. This paper proposes a practical framework for enterprise knowledge discovery in a systematical manner. The six-step framework employs the cause-andeffect diagram to model enterprise processes, tasks and attributes corresponding diagram to define data mining tasks, and multi-criteria method to assess the mined results in the form of association rules. This research also applied the proposed framework to a real case study of knowledge discovery from service records. The mining results have been proven useful in product design and quality improvement and the framework has demonstrated its applicability of guiding an enterprise to discover knowledge from historical data to tackle existing problems.