Risk assessment method and device based on XGBoost and electronic equipment

Abstract

The invention provides a risk assessment method and device based on XGBoost and electronic equipment. The method comprises steps: acquiring historical sample data, and determining positive and negative samples, wherein the historical sample data comprises user social data, user associated person data and fraud data; based on the obtained historical sample data, extracting risk characteristic variables, and establishing a training data set and a verification data set; screening risk characteristic variables according to a screening rule; constructing a risk identification model by using an XGBoost algorithm, and performing training by using the training data set; obtaining user social data and user associated person data of a current user, and calculating a risk assessment value of the userby using the risk identification model; and judging whether to trigger an early warning mechanism or not based on the risk assessment value of the user. According to the method and device, the risk condition of the user can be accurately evaluated, the fraudulent user is effectively identified, the model precision is improved, over-fitting is avoided, and the evaluation effect is remarkably improved.

Description

technical field

[0001] The present invention relates to the field of computer information processing, in particular to an XGBoost-based risk assessment method, device and electronic equipment. Background technique

[0002] Risk control (referred to as risk control) means that risk managers take various measures and methods to eliminate or reduce the various possibilities of risk cases, or risk controllers reduce the losses caused by risk cases. Risk control is generally used in the financial industry, such as risk control for company transactions, business transactions or personal transactions.

[0003] In the context of expanding business scale, diversification of fraud methods, and rising labor costs, existing fraud prevention methods tend to refine the anti-fraud management model, and establish intelligent anti-fraud management based on the fraud scoring model, forming a quantitative A business model combining analytical and qualitative analysis is imperative. Furthermo...

Images