An analysis on the use of pre-processing methods in evolutionary fuzzy systems for subgroup discovery

摘要

Subgroup discovery is a descriptive data mining technique which aims at obtaining interesting rules through supervised learning. In general, there are no works analysing the consequences of the presence of missing values in data in this task, although improper handling of this type of data in the analysis may introduce bias and can result in misleading conclusions being drawn from a research study. This paper presents a study on the effect of using the most relevant approaches for pre-processing of missing values in a determined group of algorithms, the evolutionary fuzzy systems for subgroup discovery.The experimental study presented in this paper show that, among the methods studied, the KNNI pre-processing approach for missing values obtains the best results in evolutionary fuzzy systems for subgroup discovery.