Abstract
The evaluation or metrology of commercial bank management efficiency is the core of its effective management. The existing management efficiency of commercial banks adopts evaluation scheme rather than metrology. There are four shortcomings with the evaluation scheme, the evaluation is a ranking of advantages and/or disadvantages, all objects should participate in the evaluation, the evaluation results are valid only for the evaluated objects, and the evaluation results lack the metrology benchmark of the object domain. To address this issue, the paper presents a method to determine the benchmark of management efficiency, which is similar to the unit "meter" in the metrology of "length" and the unit "scale" in the metrology of "management efficiency". Firstly, the method of cluster analysis is used to solve the problem of attribute metrology base of management efficiency metrology. Based on the master of certain attribute data of commercial bank management efficiency metrology, cluster analysis is carried out onto the attribute data, and its characteristics and patterns are mined to determine the candidate benchmark set. Secondly, in the candidate datum set, k-means method is used to determine the metrology attribute datum to obtain the general metrology attributes. Finally, the absolute metrology of management efficiency is carried out for any commercial bank according to the benchmark, and the validity and feasibility of the benchmark are verified with an example. In such a way, the deficiencies of four aspects of evaluation are solved. Such a strategy can be adapted to different banks at any different time for their respective measurement, which extends the clustering statistical methods for attribute datum determination. The results can be applied to some other fields wherein object metrology is the basic task.