How the Estonian Tax and Customs Board Evaluated a Tax Fraud Detection System Based on Secure Multi-party Computation
How the Estonian Tax and Customs Board Evaluated a Tax Fraud Detection System Based on Secure Multi-party Computation
复制标题
爱沙尼亚税务和海关委员会如何评估基于安全多方计算的税务欺诈检测系统
DOI：
--
复制
发表时间：
2015
期刊：
影响因子：
--
通讯作者：
Meril Vaht
中科院分区：
文献类型：
--
作者：
D. Bogdanov;Marko Jõemets;Sander Siim;Meril Vaht
爱沙尼亚税务和海关局（MTA）已经确定，爱沙尼亚每年因逃避增值税（VAT）而损失超过2.2亿欧元。议会提议立法，要求公司申报其购买和销售发票，以进行自动风险分析和欺诈检测。爱沙尼亚总统以违反保密规定和给公司造成不必要的负担为由否决了这项法律。在本文中，我们报告了我们与MTA合作建立一个税务欺诈检测系统原型，该原型使用安全多方计算（SMC）来消除公司对保密性的担忧。我们估计，该原型可以在10天内处理一个月的爱沙尼亚增值税数据，运行价值20000欧元的硬件。
The Estonian Tax and Customs Board (MTA) has identified that Estonia is losing over 220 million euros a year due to avoidance of value-added tax (VAT). The parliament proposed legislation that makes companies declare their purchase and sales invoices for automated risk analysis and fraud detection. The law was vetoed by the Estonian President on the grounds of confidentiality breach and unnecessary burden to companies. In this paper, we report on our collaboration with MTA to build a tax fraud detection system prototype that uses secure multi-party computation (SMC) to remove the companies’ concerns over confidentiality. We estimate that the prototype could process a month of Estonian VAT data in ten days running on 20 000 euros worth of hardware.