Risk Ordering Relation and Risk Control for P2P Lending Platforms: A Multi-Agent Systems (MAS) Approach

Abstract

In the context of Peer-to-Peer (P2P) lending, risk controls are required, and they usually refer to a set of procedures and operations that aim to protect the integrity of data, particularly for accurate financial representation within the platform. Risk control procedures need to be in place to ensure accountability and fairness in risk and return trade-offs on federated platforms. This will foster trust among participants, especially when multiple fraud cases in the past, such as Enron, Madoff Investment Securities, and WorldCom, have accentuated the importance of a robust internal control mechanism in maintaining the credibility of the financial ecosystem. Stakeholders in the P2P lending industry are becoming increasingly concerned about the issue of trust, necessitating a re-evaluation of internal control frameworks to uphold objectivity and reliability. With the growth of the P2P lending industry as an alternative lending and borrowing platform and the requirement of an autonomous P2P lending platform, complexity arises, and autonomous entities (i.e., MAS) working together to assess, monitor, and mitigate risks is the only solution for such complexities. The orchestration of MAS plays a pivotal role in facilitating and mitigating risks. This study aims to provide a process methodology for fostering collaborative dynamics within the P2P lending domain. A state diagram approach is presented, where state orders (SO), lending approvals, risk graphs, risk ordering relations, and risk bands (RB) are introduced for MAS to assume certain roles or tasks. For each task, controls for the segregation of duties are presented as well. Given the absence of proper autonomous systems for decision-making, robust internal control methods are necessary for controls to execute federated trust on lending platforms. Our approach will significantly improve investors’ confidence meant to achieve this goal. Doi: 10.28991/ESJ-2024-08-04-024 Full Text: PDF

Keywords

• p2p lending
• multi agent systems (mas)
• state order
• risk graph
• risk ordering relation.