Data Pre-processing Based on Rough Sets and the Link to Other Theories

Abstract

Data reduction is a main point of interest across a wide variety of fields. In fact, focusing on this step is crucial as it often presents a source of significant data loss. Many techniques were proposed in literature to achieve the task of data reduction. However, most of them tend to destroy the underlying semantics of the features after reduction or require additional information about the given data set for thresholding. Thus, this tutorial will be focused on presenting Rough Set Theory (RST) as a technique that can on the one hand reduce data dimensionality using information contained within the data set and on the other hand capable of preserving the meaning of the features. RST can be used as such tool to discover data dependencies and to reduce the number of attributes contained in a data set using the data alone, requiring no additional information. Basically, two main points will be discussed. First, presenting RST as a data pre-processing technique and, second, the link of RST to other theories; mainly to Fuzzy Set Theory.