A Multimodal Approach to SME Credit Scoring Integrating Transaction and Ownership Networks

2025-10-10

AI Summary

SME credit accessibility is hindered by information asymmetry and systemic risk contagion, which conventional credit scoring models fail to capture due to their inability to represent dynamic, directional risk propagation within multilayer heterogeneous networks (e.g., shared ownership, financial transactions). Method: We propose a graph neural network (GNN) that jointly encodes multilayer relational structures and structured financial data, explicitly modeling both the directionality and weighted strength of risk transmission edges. Contribution/Results: Our model enables the first interpretable, quantitative analysis of cascade default mechanisms in supply chains and similar settings. Evaluated on large-scale real-world loan data, it achieves an 8.7% AUC improvement over state-of-the-art baselines. Moreover, it supports visualizable risk-path tracing and enterprise-level risk attribution, enhancing transparency and decision support for credit underwriting.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Data Mining & Knowledge Management: Graph Mining, Social Network Analysis & Community
• Reasoning under Uncertainty: Graphical Models

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

Small and Medium-sized Enterprises (SMEs) are known to play a vital role in economic growth, employment, and innovation. However, they tend to face significant challenges in accessing credit due to limited financial histories, collateral constraints, and exposure to macroeconomic shocks. These challenges make an accurate credit risk assessment by lenders crucial, particularly since SMEs frequently operate within interconnected firm networks through which default risk can propagate. This paper presents and tests a novel approach for modelling the risk of SME credit, using a unique large data set of SME loans provided by a prominent financial institution. Specifically, our approach employs Graph Neural Networks to predict SME default using multilayer network data derived from common ownership and financial transactions between firms. We show that combining this information with traditional structured data not only improves application scoring performance, but also explicitly models contagion risk between companies. Further analysis shows how the directionality and intensity of these connections influence financial risk contagion, offering a deeper understanding of the underlying processes. Our findings highlight the predictive power of network data, as well as the role of supply chain networks in exposing SMEs to correlated default risk.

Problem

• Improving SME credit scoring using transaction and ownership network data
• Modeling default risk contagion through interconnected firm networks
• Enhancing prediction accuracy by combining network and traditional data

Innovation

• Uses Graph Neural Networks for SME credit scoring
• Integrates multilayer ownership and transaction networks
• Combines network data with traditional structured data