In-loan risk prediction method and device

Abstract

The embodiment of the invention provides an in-loan risk prediction method and device, which can be used in the technical field of artificial intelligence, and the method comprises the steps: inputting prediction data of a target user into a preset in-loan risk prediction model, and determining an in-loan risk prediction result of the target user according to the output of the in-loan risk prediction model, wherein the prediction data comprises behavior characteristic data and in-loan risk scoring data acquired by applying an in-loan behavior scoring card model; and outputting an in-loan riskprediction result of the target user. According to the invention, on the basis of ensuring the accuracy of risk prediction in loan, the acquisition efficiency and reliability of the input data for therisk prediction model in loan can be effectively improved, and the data processing difficulty and data calculation amount in the risk prediction process in loan can be effectively reduced; and thus,the intra-loan risk prediction period can be effectively shortened and the acquisition efficiency of the intra-loan risk prediction result can be improved.

Description

technical field [0001] This application relates to the technical field of data processing, in particular to the technical field of artificial intelligence, and in particular to a method and device for predicting risks in lending. Background technique [0002] In order to meet the credit risk control needs at different stages, financial institutions usually need to use the application score card before the loan, the behavior score card during the loan, and the collection score card after the loan to score the risk of financial users. The Behavior Scoring model (Behavior Scoring) is a scoring model that dynamically predicts the credit risks of financial users based on various behaviors of financial users during the use of their accounts. Since the loan-to-credit behavior scorecard model is a linear model, although its evaluation results are interpretable, due to its model nature, it has problems such as relatively high measurement accuracy and insufficient risk sensitivity. T...

AI Technical Summary

Problems solved by technology

[0004] However, although the above-mentioned loan risk prediction method can provide the accuracy of the prediction results to a certain extent, it also involves multiple The fusion process between models, so the risk prediction method in this loan requires a large amount of data calculation and takes a long time

That is to say, the existing loan risk prediction methods cannot meet the accuracy and efficiency requirements of the loan risk prediction process at the same time

Images