Granular correlation analysis in data mining

We introduce and study the use of the concept of granular correlation. Granular correlation arises as a result of introducing fuzzy information granules and can be regarded as a generic vehicle of data mining. It is shown how an analysis of fuzzy granular correlation helps reveal and quantify relationships between variables in any task of data mining. This analysis sheds light on an important issue of statistical relevance of granular associations in data sets.