A Deep Learning Approach for Credit Scoring Using Feature Embedded Transformer

In this paper, we introduce a transformer into the field of credit scoring based on user online behavioral data and develop an end-to-end feature embedded (FE-Transformer) approach. The FE-Transformer neural network is composed two parts: wide part deep part. uses network. output are concentrated in fusion layer. experimental results show that learning model proposed paper outperforms LR, XGBoost, LSTM, AM-LSTM comparison methods terms area under receiver operating characteristic curve (AUC) Kolmogorov–Smirnov (KS). This shows can accurately predict default risk.