Author(s)

J.A. Baranauskas, M.C. Monarch & G.E.A.P.A. Batista

Abstract

Classification for very large databases has many practical applications in Data Mining. Thus, Machine Learning algorithms should be able to op- erate in massive datasets. When a dataset is too large for a particular learning algorithm to be applied, there are other ways to make learning fea- sible; preprocessing techniques and dataset sampling can be used to scale up classifiers to large datasets. In this work we propose a computational environment based on two architectures, one for data pre-processing and one for post-processing which allow evaluation of induced knowledge. The two architecture share a set of learning systems, which can be enhanced to support new one

Keywords