Title: Understanding a Version of Multivariate Symmetric Uncertainty to assist in Feature Selection

Abstract: In this paper, we analyze the behavior of the multivariate symmetric uncertainty (MSU) measure through the use of statistical simulation techniques under various mixes of informative and non-informative randomly generated features. Experiments show how the number of attributes, their cardinalities, and the sample size affect the MSU. We discovered a condition that preserves good quality in the MSU under different combinations of these three factors, providing a new useful criterion to help drive the process of dimension reduction.