Camila Delefrate Martins,  Solange Oliveira Rezende. 

The data mining process aims to obtain valid, novel, useful and understandable knowledge. Therefore, it is important to develop technics to support the user when analysing the extracted knowledge, specially in the case of association, because this technique generates great volume of rules, which makes its interpretation a difficult task. Taxonomies can be used to reduce this volume, but its manual construction is a very hard work. In this paper, it is presented a method to generate taxonomies using clustering, aiming to obtain more useful and interesting taxonomies and reducing the time spent on this process.

http://www.lbd.dcc.ufmg.br/colecoes/enia/2005/025.pdf

Processando consulta.