Volume 7 : Number 1 : Paper 2

June 2004 Special Issue of Best Papers presented at JIISIC'2003. Valdivia, Chile. Guest Editor: Cecilia Bastarrica.
Title:
Contributions of KDD to the Knowledge Management Process: a Case Study

Authors and Affiliations:
Paulo de Tarso Costa de Sousa,
Hercules Antonio do Prado,
Eduardo Amadeu Moresi,
Marcelo Ladeira,

Abstract:
Knowledge Discovery in Databases (KDD), as any organizationalprocess, is carried out beneath a Knowledge Management (KM)model adopted (even informally) by a corporation. KDD isgrossly
described in three steps: pre-processing, data mining, andpost-processing. The latter is mainly related to the task oftransforming in knowledge the patterns issued in the data miningstep. On the other hand, KM comprises the following phases, inwhich knowledge is the subject of the actions: identification ofabilities, acquisition, selection and validation, organization andstorage, sharing, application, and creation. Although there aremany overlaps between KDD and KM, one of them is broadlyrecognized: the point in which knowledge arises. This paperconcerns a study aimed at clarifying relations between theoverlapping areas of KDD and knowledge creation, in KM. Thework is conducted by means of a case study using the data fromthe Electoral Court of the Federal District (ECFD), Brazil. Thestudy was developed over a 1.717.000-citizens data set fromwhich data mining models were built by applying algorithms fromWeka. It was observed that, although the importance ofInformation Technology is well recognized
in the KM realm, thetechniques of KDD deserve a special place in the knowledgecreation phase of KM. Moreover, beyond the overlap of post-processing and knowledge creation, other steps
of KDD cancontribute significantly to KM. An example is the fact that oneimportant decision
taken from the ECFD board was taken on thebasis of a knowledge acquired from the pre-processing step ofKDD.

Received 08/04/2004, Revised , Editor: Cecilia Bastarrica
Full paper, 6 pages [ PDF, 538 Kb ]