Document Type : Research Paper
Authors
1
Ph.D Candidate, Department od Economic Sciences, Finance - Banking Sciences, Qeshm Branch, Islamic Azad University, Qeshm, Iran.
2
Assistant Prof. Financial Accounting Department, Qeshm Branch, Islamic Azad University, Qeshm, Iran
3
Accounting and Financial Management, Islamic Azad University, Bandar Abbas Branch, Bandar Abbas, Iran
4
Department of Financial Management.Faculty of Management & Accounting , Eslamshahr Branch, Islamic Azad University.Islamshahr. Iran
10.22059/frj.2024.370376.1007551
Abstract
Abstract
Objective
In banks, customer management and validation is one of the vital things that is very important to maintain financial security and stability of the banking organization. One of the basic challenges in this field is to identify suitable indicators for validating and ranking customers. Due to the fact that these customers generally have limited access to financial and credit information and cannot provide guarantors or good credit records, it is very challenging to determine correct and reliable indicators. Also, in order to score customers, there is a need to determine an appropriate and fair scoring system. This system should be able to place customers in different categories by considering credit and behavioral criteria and assign them appropriate points based on their performance. In addition, there is a need to develop methods for evaluating and monitoring customers over time. The purpose of this research is identifying the indicators of validation and ranking of customers in micro-lending in the Khavar-e-Miyaneh Bank.
Methods
This research is applied-contextual in terms of purpose and exploratory in terms of method. The statistical population of this research is all retail banking customers of Digital Khavar-e-Miyaneh Bank who apply for low interest facilities of 2% annually with a penalty rate of 6%. Statistical methods in this research were carried out in two parts: descriptive and inferential statistics. In the descriptive statistics section, some personality factors such as age, gender, education, business, current debt status of the banking system, bounced checks status, money laundering status, bank account balance, bank transaction records, geographic location (place of residence and work), mobile phone model and operating system of the phone and the rating obtained from the system of Iran's credit rating consulting company and ... were analyzed and checked through tables and graphs. Naive Bayes, Meta, Attribute Selected Classifier and j48 algorithms were implemented and WEKA software was used to classify criteria and create patterns. Also, in order to evaluate the validation model and ranking of customers of unsupported microlending, T-test was used at the level of 0.25.
Findings
The findings showed that if a person applies for a loan and the status of the person's previous loans has been settled 30 days and 60 days after the due date, the amount of the facility received by this person is the maximum amount of the loan that can be paid, and the status of the person's previous loans has been settled. The higher the loan amount, the better, the person's age is above middle age, the person's degree is not a bachelor's degree, diploma or sub-diploma, the person's score is above 40, the person's phone operating system is not Android, the person's phone model is not SAMSUNG, XIAOMI, and the status If the person's money laundering approval is negative, then preferably, if necessary, the loan of that person can be approved.
Research results
In this strategy (micro-lending without support), banks should give loans to people who have full knowledge about them before entering into any new credit relationship and have collected the necessary data from them and are sure of their good credit and reputation. Banks must receive complete and comprehensive information about the borrower in such a way that this information can be relied upon, because granting facilities as much as profitability can make the bank bear the risk.
Keywords
Main Subjects
)