Scholarlink

<p>Money laundering (ML) is still a complicated research area that poses difficulties for the international financial system. Increase detection and compliance, it requires scalable, creative, and privacy-preserving solutions. In order to transform Anti-Money Laundering (AML) systems, this research offers a novel and secure framework, namely BAML, that integrates Blockchain Hyperledger Technology (BHT), particularly Hyperledger Fabric (HF), with Federated Learning (FL). On the other hand, conventional centralized approaches are not as effective because they introduce a vulnerability. The Hyperledger Fabric secure, unchangeable, and transparent ledger is used in this proposed decentralized method to record and confirm financial transactions and related verifications. While FL allows organizations to train models collaboratively and privately without sharing raw data. The creation of a dynamic chaincode-based anomaly detection tool that automates real-time transaction monitoring and the flagging of questionable activity is the main innovation of this research. With an accuracy rate of 98.33%, it makes an unprecedented detection. Together with the ability to implement and enhance detection models over distributed permissioned blockchain networks, FL A consensus-driven feedback loop is introduced by this framework, in which institutions and other relevant stakeholders work together to improve and test AML models. It guarantees ongoing education and adjustment to new machine learning strategies. In comparison to current systems, the simulation results show a 30% decrease in false positives and a 25% improvement in detection efficiency. In addition to increasing operational efficacy and detection accuracy, this type of synergy tackles important constraints including data silos, privacy protection, and legal compliance. In an increasingly linked and data-driven world, this novel framework sets a new benchmark for decentralized AML systems by fusing the benefits of BHT and FL to provide a scalable, secure, and preserving solution for contemporary financial compliance.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

BAML: a decentralized approach to secure, privacy-preserving financial compliance for enhancing anti-money laundering with blockchain hyperledger and federated learning

  • 期刊论文

  • 发表日期：

    2025年8月20日

查看全文
  • Abdullah Ayub Khan,
  • Abdulmajeed Alsufyani,
  • Nawal Alsufyani,
  • Mohamad Afendee Mohamed

摘要

Money laundering (ML) is still a complicated research area that poses difficulties for the international financial system. Increase detection and compliance, it requires scalable, creative, and privacy-preserving solutions. In order to transform Anti-Money Laundering (AML) systems, this research offers a novel and secure framework, namely BAML, that integrates Blockchain Hyperledger Technology (BHT), particularly Hyperledger Fabric (HF), with Federated Learning (FL). On the other hand, conventional centralized approaches are not as effective because they introduce a vulnerability. The Hyperledger Fabric secure, unchangeable, and transparent ledger is used in this proposed decentralized method to record and confirm financial transactions and related verifications. While FL allows organizations to train models collaboratively and privately without sharing raw data. The creation of a dynamic chaincode-based anomaly detection tool that automates real-time transaction monitoring and the flagging of questionable activity is the main innovation of this research. With an accuracy rate of 98.33%, it makes an unprecedented detection. Together with the ability to implement and enhance detection models over distributed permissioned blockchain networks, FL A consensus-driven feedback loop is introduced by this framework, in which institutions and other relevant stakeholders work together to improve and test AML models. It guarantees ongoing education and adjustment to new machine learning strategies. In comparison to current systems, the simulation results show a 30% decrease in false positives and a 25% improvement in detection efficiency. In addition to increasing operational efficacy and detection accuracy, this type of synergy tackles important constraints including data silos, privacy protection, and legal compliance. In an increasingly linked and data-driven world, this novel framework sets a new benchmark for decentralized AML systems by fusing the benefits of BHT and FL to provide a scalable, secure, and preserving solution for contemporary financial compliance.