DeepTrax: Embedding Graphs of Financial Transactions
arXiv:1907.07225
Abstract
Financial transactions can be considered edges in a heterogeneous graph between entities sending money and entities receiving money. For financial institutions, such a graph is likely large (with millions or billions of edges) while also sparsely connected. It becomes challenging to apply machine learning to such large and sparse graphs. Graph representation learning seeks to embed the nodes of a graph into a Euclidean vector space such that graph topological properties are preserved after the transformation. In this paper, we present a novel application of representation learning to bipartite graphs of credit card transactions in order to learn embeddings of account and merchant entities. Our framework is inspired by popular approaches in graph embeddings and is trained on two internal transaction datasets. This approach yields highly effective embeddings, as quantified by link prediction AUC and F1 score. Further, the resulting entity vectors retain intuitive semantic similarity that is explored through visualizations and other qualitative analyses. Finally, we show how these embeddings can be used as features in downstream machine learning business applications such as fraud detection.
References in corpus (8)
- Efficient Estimation of Word Representations in Vector Space
- Inductive Representation Learning on Large Graphs
- XLNet: Generalized Autoregressive Pretraining for Language Understanding
- struc2vec: Learning Node Representations from Structural Identity
- Representation Learning for Attributed Multiplex Heterogeneous Network
- PyTorch-BigGraph: A Large-scale Graph Embedding System
- Watch Your Step: Learning Node Embeddings via Graph Attention
- client2vec: Towards Systematic Baselines for Banking Applications