Department of Computer Engineering, Science and Research Branch, Islamic Azad University, Tehran, Iran.
Comput Intell Neurosci. 2020 Feb 8;2020:6503459. doi: 10.1155/2020/6503459. eCollection 2020.
Financial fraud is among the main problems undermining the confidence of customers in addition to incurring economic losses to banks and financial institutions. In recent years, along with the proliferation of fraud, financial institutions began looking for ways to find a suitable solution in the fight against fraud. Given the advanced and varied changes in methods of fraud, extensive research has been conducted to detect fraud. In this paper, the Artificial Neural Network technique and Harmony Search Algorithm are used to detect fraud. In the proposed method, hidden patterns between normal and fraudulent customers' information are searched. Given that fraudulent behavior could be detected and stopped before they take place, the results of the proposed system show that it has an acceptable capability in fraud detection.
财务欺诈是破坏客户信心的主要问题之一，除了给银行和金融机构造成经济损失之外。近年来，随着欺诈行为的泛滥，金融机构开始寻找在打击欺诈方面的合适解决方案。鉴于欺诈手段的先进和多样化，已经进行了广泛的研究来检测欺诈。在本文中，使用了人工神经网络技术和和声搜索算法来检测欺诈。在所提出的方法中，搜索正常和欺诈客户信息之间的隐藏模式。鉴于可以在欺诈行为发生之前检测和阻止欺诈行为，所提出系统的结果表明它在欺诈检测方面具有可接受的能力。
