Linking bank clients using graph neural networks powered by rich transactional data
Linking bank clients using graph neural networks powered by rich transactional data
复制标题
DOI：
10.1007/s41060-021-00247-3
复制
发表时间：
2020-01
影响因子：
2.4
通讯作者：
Valentina Shumovskaia;Kirill Fedyanin;I. Sukharev;Dmitry Berestnev;Maxim Panov
中科院分区：
文献类型：
--
作者：
Valentina Shumovskaia;Kirill Fedyanin;I. Sukharev;Dmitry Berestnev;Maxim Panov
金融机构获得了大量关于客户交易和资金转移的数据，这些数据可以被认为是一个随时间动态变化的大型图表。在这项工作中，我们专注于预测银行客户端网络中的新交互的任务，并将其视为一个链接预测问题。我们提出了一种新的图神经网络模型，它不仅使用网络的拓扑结构，但丰富的时间序列数据可用于图的节点和边缘。我们评估开发的方法使用的数据由一家大型欧洲银行几年。所提出的模型优于现有的方法，包括其他神经网络模型，在ROC AUC分数上的链接预测问题有显着的差距，也可以提高信用评分的质量。
Financial institutions obtain enormous amounts of data about client transactions and money transfers, which can be considered as a large graph dynamically changing in time. In this work, we focus on the task of predicting new interactions in the network of bank clients and treat it as a link prediction problem. We propose a new graph neural network model, which uses not only the topological structure of the network but rich time-series data available for the graph nodes and edges. We evaluate the developed method using the data provided by a large European bank for several years. The proposed model outperforms the existing approaches, including other neural network models, with a significant gap in ROC AUC score on link prediction problem and also allows to improve the quality of credit scoring.