{"title":"银行机器人:人工智能和机器学习驱动的银行风险管理:防止洗钱和恐怖主义融资","authors":"Alexandra Prisznyák","doi":"10.35551/pfq_2022_2_8","DOIUrl":null,"url":null,"abstract":"Based on a country study related to money laundering and terrorist financing, the Financial Action Group downgraded Hungary’s compliance with Recommendation R15 (use of new technologies). At the same time, between 2020 and 2021, the Magyar Nemzeti Bank imposed fines on several commercial banks operating in Hungary for shortcomings on complying with money laundering and terrorist financing regulations. As a gap-filling analysis, the study examines supervised (classification, regression), unsupervised (clustering, anomaly detection), and hybrid machine learning models and algorithms operating based on highly unbalanced dataset of anti-money laundering and terrorist financing prevention of banking risk management. The author emphasizes that there is no one ideal algorithm. The choice between machine learning algorithm is highly determined based on the underlying theoretical logic and additional comparative. Model building requires a hybrid perspective of the give business unit, IT and visionary management.","PeriodicalId":42979,"journal":{"name":"Public Finance Quarterly-Hungary","volume":null,"pages":null},"PeriodicalIF":0.4000,"publicationDate":"2022-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"3","resultStr":"{\"title\":\"Bankrobotics: Artificial Intelligence and Machine Learning Powered Banking Risk Management : Prevention of Money Laundering and Terrorist Financing\",\"authors\":\"Alexandra Prisznyák\",\"doi\":\"10.35551/pfq_2022_2_8\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Based on a country study related to money laundering and terrorist financing, the Financial Action Group downgraded Hungary’s compliance with Recommendation R15 (use of new technologies). At the same time, between 2020 and 2021, the Magyar Nemzeti Bank imposed fines on several commercial banks operating in Hungary for shortcomings on complying with money laundering and terrorist financing regulations. As a gap-filling analysis, the study examines supervised (classification, regression), unsupervised (clustering, anomaly detection), and hybrid machine learning models and algorithms operating based on highly unbalanced dataset of anti-money laundering and terrorist financing prevention of banking risk management. The author emphasizes that there is no one ideal algorithm. The choice between machine learning algorithm is highly determined based on the underlying theoretical logic and additional comparative. Model building requires a hybrid perspective of the give business unit, IT and visionary management.\",\"PeriodicalId\":42979,\"journal\":{\"name\":\"Public Finance Quarterly-Hungary\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":0.4000,\"publicationDate\":\"2022-01-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"3\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Public Finance Quarterly-Hungary\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.35551/pfq_2022_2_8\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q4\",\"JCRName\":\"BUSINESS, FINANCE\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Public Finance Quarterly-Hungary","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.35551/pfq_2022_2_8","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q4","JCRName":"BUSINESS, FINANCE","Score":null,"Total":0}
引用次数: 3
摘要
根据一项与洗钱和恐怖主义融资有关的国家研究,金融行动小组降低了匈牙利对建议R15(使用新技术)的遵守程度。与此同时,在2020年至2021年期间,匈牙利央行(Magyar Nemzeti Bank)对在匈牙利经营的几家商业银行处以罚款,原因是它们在遵守洗钱和恐怖主义融资法规方面存在缺陷。作为一项空白填补分析,该研究考察了监督(分类、回归)、无监督(聚类、异常检测)和混合机器学习模型和算法,这些模型和算法基于高度不平衡的反洗钱和恐怖主义融资防范银行风险管理数据集。作者强调,没有一个理想的算法。机器学习算法之间的选择在很大程度上取决于底层的理论逻辑和附加的比较。模型构建需要给定业务单元、IT和远景管理的混合透视图。
Bankrobotics: Artificial Intelligence and Machine Learning Powered Banking Risk Management : Prevention of Money Laundering and Terrorist Financing
Based on a country study related to money laundering and terrorist financing, the Financial Action Group downgraded Hungary’s compliance with Recommendation R15 (use of new technologies). At the same time, between 2020 and 2021, the Magyar Nemzeti Bank imposed fines on several commercial banks operating in Hungary for shortcomings on complying with money laundering and terrorist financing regulations. As a gap-filling analysis, the study examines supervised (classification, regression), unsupervised (clustering, anomaly detection), and hybrid machine learning models and algorithms operating based on highly unbalanced dataset of anti-money laundering and terrorist financing prevention of banking risk management. The author emphasizes that there is no one ideal algorithm. The choice between machine learning algorithm is highly determined based on the underlying theoretical logic and additional comparative. Model building requires a hybrid perspective of the give business unit, IT and visionary management.