Navigating the Dynamics of Financial Embeddings over Time

Abstract

Financial transactions constitute connections between entities and through these connections a large scale heterogeneous weighted graph is formulated. In this labyrinth of interactions that are continuously updated, there exists a variety of similarity-based patterns that can provide insights into the dynamics of the financial system. With the current work, we propose the application of Graph Representation Learning in a scalable dynamic setting as a means of capturing these patterns in a meaningful and robust way. We proceed to perform a rigorous qualitative analysis of the latent trajectories to extract real world insights from the proposed representations and their evolution over time that is to our knowledge the first of its kind in the financial sector. Shifts in the latent space are associated with known economic events and in particular the impact of the recent Covid-19 pandemic to consumer patterns. Capturing such patterns indicates the value added to financial modeling through the incorporation of latent graph representations.

DebtG: A Graph Model for Debt Relationship

“…Q. Zhan et al [10] combined knowledge graphs and machine learning to detect consumer finance fraud. A. Gogoglou et al [11] formulated the credit card transactions as a bipartite graph between account holders and the merchants they shop at. Y. Ren et al [12] utilized a bipartite graph to model the relationship between users and merchants and proposed an ensemble-based fraud detection method.…”