On Data Mining and knowledge – Questions of validity

Abstract

Understanding Data Mining (DM) as part of Information Systems (IS) this contribution examines the questions how this subordination is reasoned in a technological and business logical perspective. For this purpose certain generalisable characteristics of Enterprise Resources Planning Applications (ERP) and Management Information Systems (MIS; including here Decission Support and Expert Systems) are presented. Based on this evaluation it is asked how knowledge and DM are becoming interdependent in the field of Knowledge Management (KM) in organisations. Knowledge is here understood along the Penrose’an dichotomy of information and knowledge in the context of resources and services. The topic of knowledge validity is examined from a methodological (quantitative versus qualitative methods of research) perspective probing what key characteristics of both methods are, and how those fit into the disciplinary landscape of Organisational Studies, and organisations in more general. Tracing the inherent relationship between security and information for Penrose an alternative account of security based on Foucault is presented. In this understanding security and knowledge become devices for the process of standardisation of live in order to allow for continuation of a different abstracted object that is socially generated.

Combining arguments about validity of knowledge claims with that of security, it becomes apparent that DM based knowledge and security are formations that abstract from the human condition and attempt to constrain variability. Against that background it argued that researchers and users of DM based knowledge should be more aware of the constructed character of IS, and how much of this character is also contained in DM based knowledge due to the derived character shown in the technological and business managerial subordination.