Abstract

In this paper, it is argued that although data mining (DM) is being touted as a solution to many business problems and is basking in the glory of electronic business environments of today, as practiced currently, it reflects a preoccupation with short-run commercial applications and a neglect of the underlying theoretical issues. Although an argument can be made that theoretical precedence is not a necessary prerequisite for practical application or for commercial success, it can also be argued that an exclusive reliance on data-driven and exploratory components of pattern recognition without a corresponding attention to the causal schemas underlying patterns is destined to limit the potential for DM to evolve into a long-term solution to business problems or into an intellectual discipline in its own right. This paper presents an overview of key features and assumptions in DM and examines some of the key practical, philosophical, and foundational issues in DM