Abstract
Biclustering is a branch of data analysis, whereby the goal is to find two–dimensional subgroups in a matrix of scalars. We introduce a new approach for biclustering discrete and binary matrices on the basis of boolean function analysis. We draw the correspondence between non–extendable (maximal with respect to inclusion) exact biclusters and prime implicants of a discernibility function describing the data. We present also the results of boolean-style clustering of the artificial discrete image data. Some possibilities of utilizing basic image processing techniques for this kind of input to the biclustering problem are discussed as well.
Get full access to this article
View all access options for this article.