Ranking banks using K-Means and Grey relational method

Article history: Received January 20, 2014 Accepted 30 August 2014 Available online September 1 2014 Research has been done on evaluating the performance more emphasis either on the technical aspects of the performance evaluation or considers only one aspect of banks such as sales, revenue, etc. This paper presents an empirical investigation to measure the relative efficiency of different banks using a hybrid of K-means and Grey relational analysis. The study uses eight criteria such as resources, customer satisfaction, etc. for clustering different banks and using Grey relational analysis provides ranking of different branches of a bank named Bank Qarzol Hasane Resalat located in city of Tehran, Iran. The preliminary study has indicated that the proposed model of this paper may capable of providing some promising results. © 2014 Growing Science Ltd. All rights reserved. Clustering K-Means Grey relational Banks Ranking


Introduction
Performance measurement plays essential role on improving business units' performance and their efficiencies.During the past few years, there have been tremendous development in banking systems and the primary focus is to improve the quality of services as an objective for market retention.Performance measurement in banking industry is normally involved with various qualitative as well as quantitative criteria, which leads to the implementation of multiple criteria decision making techniques.Data mining is the result of applying sophisticated modeling techniques from the diverse fields of statistics, artificial intelligence, and database management (Yuantao & Siqin, 2008;Han & Kambert, 2001).Data mining has been widely used to determine marketing trend (Kaefer et al., 2005), customer detection (Kim & Nick Street, 2004), fraud detection (Farvares & Sepehri, 2010), etc.Today, the ability to detect the profitable customers, building a long-term loyalty in them and expanding the existing relationships is the primary key and competitive factors for a customeroriented organization.The prerequisite for having such competitive factors is the existence of a very powerful customer relationship management (CRM).The precise evaluation of customers' profitability is one of the most important reasons that lead to a successful CRM programs.RFM is a technique, which scrutinizes three properties, namely recency, frequency and monetary for each customer and scores customers based on these properties.Zalaghi and Abbasnejad Varzi (2014) presented a method, which obtains the behavioral traits of customers using the extended RFM approach and having the information associated with the customers of a firm.It then classifies the customers based on K-means algorithm and finally scores the customers in terms of their loyalty in each cluster.In their method, first the customers' records are clustered and then the RFM model items are specified through selecting the effective properties on the customers' loyalty rate based on the multipurpose genetic algorithm.Next, they are scored in each cluster based on the effect that they have on the loyalty rate.Emami and Faezy Razi (2014) presented a hybrid grey based K-means and feature selection for bank evaluation.They presented a hybrid grey relational analysis and K-means to cluster and measure the performance of banking system and applied their method for selected banks in city of Semnan, Iran.The proposed study of this paper tries to re-examine the implementation of this method for a newly established bank in city of Tehran, Iran.

K-means clustering
The K-means method has become an effective in generating good clustering results for many realworld case studies (Aggarwal et al., 2003;Jain & Dubes, 1988;Jain et al., 1999;Han & Kambert, 2001).K-means clustering is a well-known data mining clustering technique, which attempts to partition N observations into K clusters where each observation is assigned to the cluster with the closest mean.Normal evaluation of an appropriate K is executed by minimizing the inner-cluster variation and maximizing the among-cluster variation, concurrently.K-means clustering is often sensitive to outliers, therefore, outliers have to be removed before accomplishing clustering (Ying & Feng, 2008;Cheng & Chen, 2008;Farvaresh & Sepehri, 2010).Edwards (2003) and Kantardzic (2011) described the K-means method with the following steps, 1. Select a primary part of K categories including samples randomly selected and compute the mean of each pair, 2. Generate a new section of each part by computing the nearest center core, 3. Measure the new batches as the main centers, 4. Repeat step 2 and step 3 until the algorithm approaches termination criteria.

Grey Relational Analysis
Grey relation analysis used in this paper follows the following steps (Deng, 1989;Hsia et al., 2004;Huang et al., 2008;Razi et al., 2013): Let X 0 be the reference and N alternatives with k criteria as follows, Grey relational coefficient are computed as follows, where 0i X  is the absolute difference between X 0 and X i in k th criterion, 0i . The computations of Grey relational degree is as follows, where w j is the weight of criterion j and we apply 1 j W k  .Finally, all relationships have to be normalized as follows, * ( ) min ( ) ( ) max ( ) min ( ) Grey relational analysis has been widely applied in different industries.Gupta and Kumar (2013), for example, used optimization of performance characteristics in unidirectional glass fiber reinforced plastic composites by applying Taguchi method and Grey relational analysis.Performance characteristics such as surface roughness and material removal rate in this paper were optimized using rough cutting operation.Salardini (2013) implemented analytical hierarchy process and grey relational analysis to offer a method for portfolio management.They implemented a statistical sample with 16 firms and applied analytical hierarchy process as well as gray relational analysis to assign weight to each firm.
The proposed study of this paper applies a hybrid of Grey relational analysis as well as K-Means for clustering 26 banks in city of Tehran, Iran based on different criteria.

The results
In this section, we present some details of our results on clustering 26 banks based on 24 different criteria defined in Table 1 as follows, The amount of resources

K-Means implementation
The implementation of K-Means has been accomplished on Clementine®12 yields two clusters shown in Table 2 and Table 3 as follows,

GRA results
The implementation of grey relational analysis has been performed on two clustering data and Table 4 and Table 5 shows the summary of ranking.

Conclusion
In this paper, we have measured the relative efficiencies of 26 branches of Bank Qarzol Hasane Resalat in city of Tehran, Iran.The proposed study has applied K-means clustering along with gray relational analysis for ranking various branches of Bank Qarzol Hasane Resalat.The implementation of K-Means clustering method has yielded two clusters and they were ranked using gray relational method.The preliminary results of this survey have indicated that the proposed study of this paper may provide a quick approach for ranking various similar units.

Table 1
The criteria used for clustering banks

Table 2
The summary of standard data for the first cluster

Table 3
The summary of standard data for the second cluster

Table 4
The summary of ranking for the first cluster

Table 5
The summary of ranking for the second cluster