Application of Data Mining and Machine Learning Techniques to Predict Loan Approval and Payment Time | ||
| Journal of Modeling and Simulation in Electrical and Electronics Engineering | ||
| دوره 4، شماره 1 - شماره پیاپی 15، تابستان 2024، صفحه 19-28 اصل مقاله (481.44 K) | ||
| نوع مقاله: Research Article | ||
| شناسه دیجیتال (DOI): 10.22075/mseee.2025.35663.1183 | ||
| نویسندگان | ||
| Ehsan Allah khoshkhoy Nilash1؛ Mansour Esmaeilpour* 2؛ Behrooz Bayat3؛ Alireza Isfandyari Moghaddam4؛ Erfan Hassannayebi5 | ||
| 1PhD Student of Information Technology Management, Department of Management, Hamedan Branch, Islamic Azad University, Hamedan, Iran. | ||
| 2Associate Professor, Department of Computer Engineering, Hamedan Branch, Islamic Azad University, Hamedan, Iran. | ||
| 3Assistant Professor, Department of Knowledge and Information Science, Hamedan Branch, Islamic Azad University, Hamedan, Iran. | ||
| 4Professor, Department of Knowledge and Information Science, Hamedan Branch, Islamic Azad University, Hamedan, Iran. | ||
| 5Assistant Professor, Department of Industrial Engineering, Sharif University of Technology, Tehran, Iran. | ||
| چکیده | ||
| One of the most important issues regarding banks is knowing the customers, their behaviors, and the decisions these institutions make regarding customers' preferences. Their main task is to provide banking facilities. Bank facilities carry the risk of default in repayment. Failure to evaluate and review factors related to repayment can cause significant damage to banks. On the other hand, investment in the private sector and various industries is also increasingly important. This action can lead to economic growth, increased employment, and national income. This research aims to identify the effective features related to the fixed capital facility data of one of the active banks in Iran, in line with the classification of customers into two categories good customers and overdue customers to predict the duration of the facility payment. The five-step method is based on data mining techniques. The most important steps of this method are data preparation, analysis with rough set methods, and common classification techniques such as artificial neural networks, tree types, Bayes types, and support vector machines. One of the most important results of this research was the identification of the features that affect the repayment and duration of fixed capital facilities. Additionally, among other results of the present research, the ANN method demonstrated superior performance in evaluating credit risk with an accuracy value of 70.27%, and the J48 technique showed superior performance in predicting the duration of payment of facilities with an accuracy of 72.54%. | ||
| کلیدواژهها | ||
| Fixed capital facility؛ Facility repayment؛ Credit risk؛ Data mining؛ Classification | ||
| مراجع | ||
|
| ||
|
آمار تعداد مشاهده مقاله: 420 تعداد دریافت فایل اصل مقاله: 558 |
||
| تعداد نشریات | 22 |
| تعداد شمارهها | 721 |
| تعداد مقالات | 10,358 |
| تعداد مشاهده مقاله | 72,736,278 |
| تعداد دریافت فایل اصل مقاله | 64,384,924 |