ABSTRACT
Microarray data analysis is an essential step for the applications mentioned above in order to extract relevant biological knowledge embedded in these large datasets. However, the process of knowledge extraction is not a trivial task and data mining techniques are often used for this purpose. Among the techniques used, we can cite clustering [3] where we assume that the genes in a group may have similar behavior under all experimental conditions. Furthermore, clustering tries to find nonoverlapping groups of genes [4]. Another interesting technique, called biclustering [5, 6], allows the identification of groups of genes that exhibit coherent expression forms across specific groups of experimental conditions. Biclustering relates to a different group of clustering algorithms that perform simultaneous row and column clustering. Biclustering algorithms have also been used in the literature to refer to other application fields with different names such as co-clustering, bi-dimensional clustering, and subspace clustering [6].
