Small and medium enterprises (SME) are crucial for economy and have a higher exposure rate to default than large corporates. In this work, we address the problem of predicting the default of an SME. Default prediction models typically only consider the previous financial situation of each analysed company. Thus, they do not take into account the interactions between companies, which could be insightful as SMEs live in a supply chain ecosystem in which they constantly do business with each other. Thereby, we present a novel method to improve traditional default prediction models by incorporating information about the insolvency situation of customers and suppliers of a given SME, using a graph-based representation of SME supply chains. We analyze its performance and illustrate how this proposed solution outperforms the traditional default prediction approaches.

Credit scoring is an important tool to assess the solidity of small and medium-sized enterprises (SMEs), and to unlock for them new options for credit and improvement of cash flow. Credit scoring is, in its most common form, used by (potential) creditors to predict the probability of SMEs to default in the future, as an inverse measure of creditworthiness. The majority of existing credit scoring methods for SMEs are solely based on the analysis of SMEs’ financial data. While straightforward, these methods have major limitations: they may rely on very incomplete or outdated data, and fail to capture the very dynamic environment in which the business of SMEs evolves. In this paper, we propose an alternative approach to credit scoring for SMEs by enriching traditionally used financial data with social media data. We carried out our analysis on 25654 SMEs in the Netherlands, using 20 traditional financial indicators and 35 social media features. Experimental results suggest that the use of social media data in addition to traditional data significantly improves the quality of the credit scoring model for SMEs. Furthermore, we analyze the most important factors from social media data influencing the credit scoring.