Non-parametric Methods for Correlation Analysis in Multivariate Data with Applications in Data Mining

In this thesis, we develop novel methods for correlation analysis in multivariate data, with a special focus on mining correlated subspaces. Our methods handle major open challenges arisen when combining correlation analysis with subspace mining. Besides traditional correlation analysis, we explore interaction-preserving discretization of multivariate data and causality analysis. We conduct experiments on a variety of real-world data sets. The results validate the benefits of our methods.