An intelligent tool for expediting and automating data mining steps

Abstract

Data mining features a variety of techniques and algorithms that can draw interesting conclusions, utilizing large volumes of data. However, the data mining process itself is a cumbersome task, which requires expert knowledge, concerning preprocessing of data, selection of algorithms, appropriate configuration, feature identification and result comprehension. In this paper, the design and development of an expert system attempting to expedite and automate the aforementioned steps is presented and discussed. The resulting prototype platform aims to provide data mining capabilities to users that are unfamiliar with the field of data mining, while at the same time it features a high degree of customization and advanced features, which render it useful to data mining experts for easily experimenting and comparing different configurations and algorithms. For testing and demonstration purposes, the proposed platform is employed on a case study concerning actual medical data.