As Wolpert and Macready [4]pointed out in their No Free Lunch (NFL) theorem that there exists no single algorithm that could achieve the best performance for all measures for a given problem domain. Thus, a list of data mining algorithm ranking is more useful than providing the best performed algorithm for a particular task [5]. In addition, the preferences of users play an important role in algorithms evaluation and selection. One way to get users involved in the algorithm selection procedure is to allow them to assign priorities to performance measures. Since MCDM methods can satisfy both requirements, they have great potential in the area of ranking data mining algorithms.