Bank card fraud detection using artificial neural network
Publish Year: 1394
نوع سند: مقاله ژورنالی
زبان: English
View: 197
This Paper With 26 Page And PDF Format Ready To Download
- Certificate
- من نویسنده این مقاله هستم
این Paper در بخشهای موضوعی زیر دسته بندی شده است:
استخراج به نرم افزارهای پژوهشی:
شناسه ملی سند علمی:
JR_JITM-6-4_010
تاریخ نمایه سازی: 26 بهمن 1400
Abstract:
There is no accurate data for the bank cards fraud in Iran. But, it seems to be a growing trend in this regard and in the near future it is going to become one of the critical problems in Iran's banking system. Unfortunately, not enough research works have been done in this field in our country and the banking system requires models that are efficient enough to ensure safe use of bank cards. In this paper, after identifying the most common types of bank cards frauds and fraudulent transactions simulation, Artificial Neural Network (ANN) was used for the classification of transactions into two types of legitimate (non-fraud) and fraudulent (suspicious) actions. The proposed model is a Multi-Layer Perceptron (MLP) neural network designed based on the domestic banking system and is able to classify the transactions with more than ۹۹ percent accuracy. Measures of performance calculated in this study are compared with the results of other research models. The results show that the proposed model is quite reliable and valid.
Keywords:
Authors
ملیحه وثوق
MSc. in Information Technology Management, Science and Research Branch, Islamic Azad University, Tehran, Iran.
محمد تقی تقوی فرد
Assistant Prof., Allameh Tabatabaii University, Tehran, Iran.
محمود البرزی
Assistant Prof., Science and Research Branch, Islamic Azad University, Tehran, Iran.
مراجع و منابع این Paper:
لیست زیر مراجع و منابع استفاده شده در این Paper را نمایش می دهد. این مراجع به صورت کاملا ماشینی و بر اساس هوش مصنوعی استخراج شده اند و لذا ممکن است دارای اشکالاتی باشند که به مرور زمان دقت استخراج این محتوا افزایش می یابد. مراجعی که مقالات مربوط به آنها در سیویلیکا نمایه شده و پیدا شده اند، به خود Paper لینک شده اند :