Business data risk classification threshold determination method and device, equipment and storage medium

Abstract

The embodiment of the invention provides a business data risk classification threshold determination method and device, equipment and a storage medium. The method comprises the steps of obtaining a risk probability prediction value of historical business data and a real occurrence probability of a dangerous event; dividing the risk probability prediction value into a plurality of interval groups according to the size; performing monotonicity processing on the plurality of interval groups to obtain a plurality of interval groups conforming to monotonicity; and classifying the plurality of interval groups conforming to the monotonicity according to the real occurrence probability of the dangerous event, and determining a risk classification threshold according to a classification result so as to be used for business risk prediction. According to the embodiment of the invention, the accuracy and efficiency of setting the business data risk classification threshold can be improved.

Description

technical field [0001] This description relates to the technical field of data processing, and in particular to a method, device, device and storage medium for determining a business data risk classification threshold. Background technique [0002] Risk classification and identification are involved in many business areas. For example, the classification and identification of production safety risks in the industrial field; the classification and identification of credit loan risks in the financial field; the classification and identification of network and information security risks in the Internet field; the classification and identification of health risks in the medical and insurance fields, etc. [0003] Before performing business risk classification and identification, it is necessary to determine the business data risk classification threshold. In the prior art, the business data risk classification threshold is generally set manually based on experience. However, i...

AI Technical Summary

Problems solved by technology

However, in many cases, the manually set business data risk classification threshold is not necessarily accurate. When using the inaccurate business data risk classification threshold as a reference to classify and identify business data, it is difficult to obtain the real risk of business data.

Not only that, for each application scenario, it is necessary to manually set the corresponding risk classification threshold, and the efficiency of setting the business data risk classification threshold is low

Images