In the field of finance, the prediction of individual credit default is of vital importance. However, existing methods face problems such as insufficient interpretability and transparency as well as limited performance when dealing with high-dimensional and nonlinear data. To address these issues, this paper introduces a method based on Kolmogorov-Arnold Networks (KANs). KANs is a new type of neural network architecture with learnable activation functions and no linear weights, which has potential advantages in handling complex multi-dimensional data. Specifically, this paper applies KANs to the field of individual credit risk prediction for the first time and constructs the Kolmogorov-Arnold Credit Default Predict (KACDP) model. Experiments show that the KACDP model outperforms mainstream credit default prediction models in performance metrics (ROC_AUC and F1 values). Meanwhile, through methods such as feature attribution scores and visualization of the model structure, the model's decision-making process and the importance of different features are clearly demonstrated, providing transparent and interpretable decision-making basis for financial institutions and meeting the industry's strict requirements for model interpretability. In conclusion, the KACDP model constructed in this paper exhibits excellent predictive performance and satisfactory interpretability in individual credit risk prediction, providing an effective way to address the limitations of existing methods and offering a new and practical credit risk prediction tool for financial institutions.

The correlation-based financial networks, constructed with the correlation relationships among the time series of fluctuations of daily logarithmic prices of stocks, are intensively studied. However, these studies ignore the importance of negative correlations. This paper is the first time to consider the negative and positive correlations separately, and accordingly to construct weighted temporal antinetwork and network among stocks listed in the Shanghai and Shenzhen stock exchanges. For (anti)networks during the first 24 years of the 21st century, the node's degree and strength, the assortativity coefficient, the average local clustering coefficient, and the average shortest path length are analyzed systematically. This paper unveils some essential differences in these topological measurements between antinetwork and network. The findings of the differences between antinetwork and network have an important role in understanding the dynamics of a financial complex system. The observation of antinetwork is of great importance in optimizing investment portfolios and risk management. More importantly, this paper proposes a new direction for studying complex systems, namely the correlation-based antinetwork.