Biclustering methods have been initially developed for solving tasks of finding local correlations between expressions of gene subsets in the subsets of conditions. Later on they started to be employed in target marketing for revealing preferences of subsets of customers/buyers over the subsets of products/services. It can be stated with confidence that in the future these methods will find a wide application in other research areas for mining knowledge when initial data are of specific character. This paper provides a short description and analysis of the four well-known biclustering methods in the order of their evolution.