反洗钱机器学习_为什么不使用机器学习来打击洗钱活动？

反洗钱机器学习

An interesting article in the MIT Tech Review, “The pandemic has changed how criminals hide their cash — and AI tools are trying to sniff it out”, describes how the coronavirus pandemic, with its lockdowns, travel disruption and restrictions on economically or cash-sensitive activities, is a headache for criminals trying to launder money.

麻省理工学院技术评论中的一篇有趣的文章，“ 大流行改变了罪犯隐藏现金的方式，而AI工具正试图嗅出它 ”，描述了冠状病毒的大流行及其封锁，旅行中断以及对经济或现金的限制。敏感的活动，对于试图洗钱的罪犯来说是头疼的事情。

Each year, according to UN estimates, between $800 billion and $2 trillion of criminal proceeds are laundered and reintroduced into the economy, usually using small businesses that primarily handle cash, and moving it illegally across borders. But the restrictions imposed by the pandemic have forced criminals to seek new strategies to launder their ill-gotten gains that are proving difficult for the authorities to tackle.

据联合国估计 ，每年通常有8000亿至2万亿美元的犯罪收益被洗掉并重新引入经济，通常使用主要是处理现金并非法跨境转移的小企业。 但是，由这种大流行所施加的限制迫使犯罪分子寻求新的战略来洗刷他们的不义之财，事实证明，这是当局难以解决的。

The solution could lie in using machine learning’s ability for anomaly detection. Applied to a set of data, anomaly detection identifies outliers that may indicate fraud or data quality problems, without the need to have previously labelled the data. Unsupervised learning tools for anomaly detection — those I am most familiar with are developed by BigML, a company to which I am a strategic advisor — assign a score to each case in a data set of between 0% and 100%, and scores of 60% or more are usually considered outliers.

解决方案可能在于利用机器学习的异常检测能力。 将异常检测应用于一组数据后，无需事先标记数据即可识别出可能指示欺诈或数据质量问题的异常值。 我最熟悉的无监督学习工具- 由我是战略顾问的公司BigML开发 -我最熟悉的工具是在0％至100％的数据集中为每个案例分配分数，分数为60通常认为％或更高的异常值。

In many cases, detecting, isolating and eliminating outliers is done systematically to improve the quality of a data set before analysis to find mis-entered information, problems with reading instruments, etc., and can lead to significant improvements in the accuracy and performance when assessing classification and regression models. But in other cases, such as banking or credit card transactions, cyber security, etc., this type of analysis can be used to identify patterns of fraud or intrusion, and can be applied very quickly and easily.

在许多情况下，系统地检测，隔离和消除异常值是为了提高数据集的质量，然后再进行分析以查找错误输入的信息，阅读工具存在的问题等，并可以显着提高准确性和性能。评估分类和回归模型。 但是在其他情况下，例如银行或信用卡交易，网络安全等，这种类型的分析可以用来识别欺诈或入侵的模式，并且可以非常快速，轻松地应用。

Soon, we will be able to subject data to such analyses at multiple control points. In politics, for example, this kind of technology can detect and isolate cases of corruption, which tend to generate irregular patterns. In the economy at large, the progressive reduction in the use of cash and the consequent analysis of the time series of electronic transactions will also make possible more immediate and direct ways of detecting patterns of money laundering or tax evasion, prompting criminals to seek more sophisticated methods to carry out their activities.

很快，我们将能够在多个控制点对数据进行此类分析。 例如，在政治中，这种技术可以发现并隔离容易产生不规则模式的腐败案件。 在整个经济中，现金使用量的逐步减少以及对电子交易时间序列的分析也将使人们有更多直接和直接的方式来检测洗钱或逃税的方式，从而促使犯罪分子寻求更复杂的手段。开展活动的方法。

Illicit money is a huge problem that has major social and economic repercussions. For criminals, the pandemic has proved a perfect storm when it comes to blocking their usual sources of laundering, which also coincides with a greater availability of simple analytical tools for detecting their activity patterns. Can we count on the political will needed to put such systems in place? Would this not be a good argument for increasing the adoption of machine learning tools by the authorities?

非法货币是一个巨大的问题，具有重大的社会和经济影响。 对于犯罪分子而言，在阻止其通常的洗钱来源方面，这种流行病已被证明是一场完美的风暴，这也与用于检测其活动模式的简单分析工具的可用性有所提高。 我们能否指望建立这样的系统所需的政治意愿？ 这是否不是增加当局使用机器学习工具的好理由？

(En español, aquí)

( Enespañol ， aquí )

翻译自: https://medium.com/enrique-dans/why-not-use-machine-learning-to-crack-down-on-money-laundering-43c025573378

反洗钱机器学习