Optimizing Transactional Privacy and Regulatory Compliance in Decentralized Finance: A Machine Learning Approach Using Hyperledger Fabric

Kautaman, Gayuh (2023) Optimizing Transactional Privacy and Regulatory Compliance in Decentralized Finance: A Machine Learning Approach Using Hyperledger Fabric. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 05111942000002-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 October 2025. Download (1MB) | Request a copy

Abstract

Seiring pertumbuhan pesat bidang Keuangan Terdesentralisasi (DeFi), tantangan untuk memastikan privasi dan keamanan transaksi sekaligus mematuhi regulasi Know-Your- Customer (KYC) menjadi semakin signifikan. Kebutuhan akan pendekatan yang dapat menyeimbangkan persyaratan yang tampaknya bertentangan ini sangat mendesak, karena kegagalan dalam hal ini dapat menghambat penerimaan dan penggunaan sistem DeFi secara lebih luas. Studi ini mengatasi masalah ini dengan mengusulkan solusi unik yang didorong oleh teknologi, yang mengintegrasikan kemampuan data pribadi Hyperledger Fabric dan model pembelajaran mesin. Metodologi penelitian ini berputar di sekitar sistem otonom untuk verifikasi identitas pelanggan dan deteksi aktivitas ilegal berdasarkan riwayat transaksi, sehingga mempertahankan prinsip dasar privasi pelanggan. Secara khusus, penelitian ini mengevaluasi riwayat transaksi Ethereum, sementara identitas pelanggan dinilai di blockchain pribadi dan berizin. Proses KYC dilakukan di mesin lokal pelanggan menggunakan kontrak pintar dan data pribadi. Model pembelajaran mesin digunakan untuk mengidentifikasi pola dan menandai transaksi yang mencurigakan. Pendekatan inovatif ini bertujuan untuk meningkatkan keamanan secara keseluruhan dan mengurangi redundansi dalam proses KYC. Metodologi yang diusulkan menunjukkan janji tidak hanya dalam menjaga privasi pelanggan tetapi juga dalam mengurangi secara signifikan biaya dan waktu regulasi yang terkait dengan proses KYC tradisional. Temuan dari studi ini menunjukkan bahwa pendekatan ini dapat mendorong lingkungan yang lebih efisien, aman, dan patuh bagi pengguna DeFi dan institusi keuangan. Implikasi dari temuan ini menandai kontribusi signifikan untuk bidang DeFi dan regulasi keuangan, dan mereka menunjukkan potensi untuk standar baru prosedur KYC di era keuangan digital.
================================================================================================================================
As the field of Decentralized Finance (DeFi) continues to experience rapid growth, the challenge of ensuring transactional privacy and security while also adhering to Know-Your- Customer (KYC) regulations becomes increasingly significant. The need for an approach that can balance these seemingly conflicting requirements is urgent, as failure to do so could hinder the broader acceptance and use of DeFi systems. This study addresses this problem by proposing a unique, technology-driven solution that integrates the private data capabilities of Hyperledger Fabric and machine learning models. The methodology of this research revolves around an autonomous system for customer identity verification and illicit activity detection based on transaction history, thereby maintaining the fundamental principle of customer privacy. Specifically, the study evaluates Ethereum transaction histories, while customer identities are assessed on a private and permissioned blockchain. The KYC process is executed on the customer's local machine using smart contracts and private data. Machine learning models are employed to identify patterns and flag suspicious transactions. This innovative approach aims to enhance the overall security and reduce redundancy in the KYC process. The proposed methodology shows promise not only in preserving customer privacy but also in significantly reducing the regulatory costs and time associated with traditional KYC processes. The findings of this study suggest that this approach can promote a more efficient, secure, and compliant environment for DeFi users and financial institutions alike. The implications of these findings mark a significant contribution to the field of DeFi and financial regulation, and they indicate the potential for a new standard of KYC procedures in the digital finance era.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Anonymity, Blockchain, Computer Vision, Hyperledger Fabric, KYC, Privacy, Smart Contract, Transactions, Machine Learning, Decentralized Finance, Ethereum, Identity Verification, Illicit Activity Detection, Autonomous System, Digital Finance, Financial Regulation.
Subjects:	Q Science > Q Science (General) > Q325.5 Machine learning. Support vector machines. Q Science > QA Mathematics > QA76.9.A25 Computer security. Digital forensic. Data encryption (Computer science)
Divisions:	Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Informatics Engineering > 55201-(S1) Undergraduate Thesis
Depositing User:	Gayuh Kautaman
Date Deposited:	21 Jul 2023 04:16
Last Modified:	21 Jul 2023 04:16
URI:	http://repository.its.ac.id/id/eprint/98802

Actions (login required)

View Item