IJCATR Volume 7 Issue 1

Presenting a Model for Identifying the Best Location of Melli Bank ATMS by Combining Clustering Algorithms and Particle Optimization

Abdolhussein Shakibayinia , Faraz Forootan
10.7753/IJCATR0701.1002
keywords : Particle optimization algorithm, Banking, ATM

PDF
The Interbank Information Exchange Network (Shetab or Acceleration) has started since 2002 and the purpose was integrating and connecting card systems of all banks in the country. Currently, the Acceleration Center has been acting as Melli bank card switch in the country, and all the banks in the country are its member. These operations cover a wide range of transactions, such as cash withdrawals, electronic purchases, fund transfers, paying bills and residual payments. Shetab center processes more than two and a half million transactions per day. At present, the amount of fees received from each network transaction is 500 to 22,000 Rial, which is considered as a fee for the client's bank as revenue and for the client bank. And it does not cost any expenses to the customer, thus banks are looking for earning revenue from this service. In this, first the list of ATMs that Melli Bank pays them service fee are considered, then by using the clustering algorithm, locations were arranged for an ATM so Melli Bank pay less fee. In this study, the combination of three K-means algorithms and particle optimization algorithm and genetic algorithm were used. Davies-Bouldin Index was used to assess clustering. Then, the proposed clustering along with another clustering algorithm was evaluated and it was shown that the proposed algorithm is performing better. 8 locations for ATM were presented in proposed clustering algorithm, which is the result of the proposed clustering.
@artical{a712018ijcatr07011002,
Title = "Presenting a Model for Identifying the Best Location of Melli Bank ATMS by Combining Clustering Algorithms and Particle Optimization",
Journal ="International Journal of Computer Applications Technology and Research(IJCATR)",
Volume = "7",
Issue ="1",
Pages ="12 - 18",
Year = "2018",
Authors ="Abdolhussein Shakibayinia , Faraz Forootan"}
  • T In this research examined Clustering Bank Customers By K-means
  • Data mining, decision trees and neural network techniques have been investigated
  • In the simulation, a performance evaluation of the algorithm is carried out
  • In this research the criteria used to assess the accuracy.