Research on Personal Credit Evaluation Based on Characteristic Contribution Degree

Abstract

With the construction of China’s social credit system and the continuous improvement of the credit system, the field of personal credit has also been rapid development. The evaluation of personal credit based on mobile telecommunication data is one of the hotspots of current research. However, due to the complexity and diversity of personal credit evaluation variables, in order to reduce the complexity of the model and improve the prediction accuracy of the model, we need to reduce the dimension of the input variables. Based on the data provided by operators, this paper selects two different types of methods, namely feature extraction (principal component analysis method) and feature selection (woe-iv method) to select variables, using logistic regression method to establish the statistical analysis model of personal credit evaluation. The model can be used to evaluate the risk of personal credit, distinguishing “good” customers and “bad” customers, effectively provide technical reference and decision-making basis for operators, and realize precision marketing.