Identifying criteria which improve efficiency in an Iranian development bank using artificial neural networks

Banks, in general, have a direct impact on the macro-economy of all countries. Recognizing the criteria which have momentous influence on bank branches’ efficiency is the main purpose of this research. An artificial neural network approach, one of the most applicable data mining techniques, is adopted to identify the criteria that influence the branches' efficiency the most (according to the result of efficiency evaluation base on MCDM). Then, the optimal group of input criteria is determined in order to achieve the most efficient performance. Branches that enjoy more appropriate inputs would have better conditions to increase their efficiency, possess more acceptable position and gain more adequate results. In this paper, utilizing data mining science, we have endeavored to suggest a suitable method in recognizing the most significant inputs with positive impact on enhancing efficiency of branches by the incorporation of relatively neglected indicators which fit the particular conditions of Iranian banks. The strength of this article compared to other related researches is that it provides a mechanism according to which senior managers in the banking sector will be able to identify the most important indicators and implement the best conditions to achieve the highest level of efficiency in the collection.