Data Mining Tools and their Role in Knowledge Management

Knowledge Management (KM) as a field of study in the social technology space is driven both by the practical needs of organisations and interactions between related broad areas including cognitive sciences, information sciences, economics and management sciences (Suresh et al. 2006). Technologies play a key role in delivering and supporting KM services, and this paper will focus on knowledge discovery in relation to the process of knowledge creation in the industrial setting. As a subset of Knowledge Discovery in Databases (KDD), data mining has been defined as the "nontrivial extraction of implicit, previously unknown and potentially useful information from data". Two strands of KM are identifying existing knowledge and creating new knowledge ? data mining offers organisations the facilities to discover, organise, check and analyse their body of knowledge. Data Mining (DM) tools use data to build a model of the real world and the result of this modelling is a description of patterns and relationships in data, which can be used in pursuit of the primary data mining goals, i.e. prediction and description. Describing patterns and relationships in a complex dataset can provide the knowledge that guides future business actions. There are a range of data mining techniques for dealing with large-scale databases and sophisticated algorithms are incorporated into commercial software. This paper brings together some existing frameworks and schemes to present a set of criteria for DM tools with the aim of assisting industrial users and researchers in the selection process. PolyAnalyst from Megaputer Intelligence is used as a case study here to highlight the approach to evaluation of functionality and usability in the context of the particular business goal.