Federated Risk-Based Access Control Model for P2P Lending Platforms: A Multi-Agent Systems (MAS) Approach

Abstract

This study addresses the inherent risk management challenges in decentralized finance, particularly for peer-to-peer (P2P) lending platforms. We propose a novel framework that leverages a Multi-Agent System (MAS) to establish a collaborative network encompassing loan originators, investors, regulators, and service providers. This distributed approach facilitates federated risk management, where risk assessment and mitigation responsibilities are shared across these entities. The MAS employs a comprehensive nine-factor assessment (detailed in Table 5) to evaluate industry risk profiles, considering industry environment, competition, and internal capabilities. This data is further visualized using a color matrix (Tables 5 & 6) and utilized alongside state diagrams (Figure 2) to depict the workflow and manage tasks within the P2P lending process. Additionally, the MAS informs a novel Federated Risk-Based Access Control (FRkBAC) system that tailors access permissions (lending origination, disbursement, etc.) based on dynamic risk assessments of industry trends and individual borrower profiles. This data-driven approach fosters trust within the P2P ecosystem and represents a significant advancement in decentralized finance risk management compared to traditional methods. Doi: 10.28991/ESJ-2024-08-06-05 Full Text: PDF