Deep Learning in Consumer Credit Assessment

With the rapid development of machine learning techniques, the combination of deep learning models and causal inference methods becomes increasingly important in classification tasks. In this paper, we propose a framework that combines the Copula PC (CoPC) algorithm and deep neural networks (DNNs) for processing and analysing causality and classification tasks in mixed data. The CoPC algorithm improves the accuracy and robustness of causal inference by introducing the Gaussian Copula model and the concept of effective number of data points, especially when processing mixed data. In addition, this paper uses the deep learning model to classify the data after causal inference and test the classification results. The experimental results show that the proposed method performs well on multiple evaluation indicators such as Accuracy, Precision, Recall, and F1.In particular, in the context of high data complexity, the model in this paper has better performance and generalization ability than traditional causal inference and classification methods. The results of this paper can not only play a new role in the field of credit evaluation, but also provide reference for causal inference and classification tasks in other fields.