Investigating the effective factors in measuring customers' credibility with a combined approach of data mining and multidisciplinary decision making Problem

Changiz Delivand, Maryam; Bameni Moghadam, Mohammad; Shamami, Nader; Taghipour, Mohammad

doi:10.22034/jmor.2024.242893

Investigating the effective factors in measuring customers' credibility with a combined approach of data mining and multidisciplinary decision making Problem

Document Type : Original Article

Authors

Maryam Changiz Delivand ¹

Mohammad Bameni Moghadam ²

nader shamami ³

Mohammad Taghipour ⁴

¹ MasterofIndustrialEngineering,IslamicAzadUniversity,Tehran,Iran

² Professor, Department of Statistics, Allameh Tabatabai University, Tehran,Iran

³ Higher Research Institute of War- University of Command and Staff AJA

⁴ Department of Industrial Engineering, Ooj Institute of Higher Education, Qazvin, Iran

10.22034/jmor.2024.242893

Abstract

In order to provide loans and facilities, banks must be able to identify and classify their customers based on their ability to repay on time. In this way, banks can achieve the least risk and the highest return. This research tries to provide a better opportunity for banks to identify their customers by using two techniques of data mining and multi-indicator decision making. The statistical population of this study is all customers of 98 branches of the National Bank of West Tehran since 1396 (207104 people) who have deposits of more than ten million Rials. In this research, the effective parameters in credit risk are classified according to their importance and using the multi-criteria decision-making method, the customer’s request the facility and the effective factors in measuring the credit of the bank's customers obtained by AHP method include: history of cooperation with the bank. Debt history, loan amount, GPA, customer capital, type of ownership, place of work-living, annual income of the applicant, one-year bank account GPA, loan interest rate, loan term, current capital flow and current capital. Findings show that the average turnover index is the first priority and the applicant's annual income is the last priority.

Keywords

Credit Risk

Validation

Data Mining

Clustering

Multi-Index Decision Making