Similarity-based evaluation method, apparatus and device, and readable storage medium

Abstract

The invention discloses a similarity-based evaluation method. The method comprises the steps of obtaining first information data of a to-be-evaluated object, second information data of an evaluated first object, a weight value corresponding to associated information data and third information data of an evaluated second object; calculating a first relationship similarity between the to-be-evaluated object and an evaluated second object and a second relationship similarity between the evaluated first object and the evaluated second object according to the weight values corresponding to the first information data, the second information data, the third information data and the associated information data; and according to the first relationship similarity and the second relationship similarity, calculating a third relationship similarity between the to-be-evaluated object and the evaluated first object by using a SimRank algorithm. The invention further discloses a similarity-based evaluation device and equipment and a readable storage medium. According to the invention, gang financial fraud can be effectively assessed based on the similarity of the information data, and the accuracyof credit assessment is improved.

Description

technical field [0001] The invention relates to the technical field of credit evaluation, in particular to an evaluation method, device, equipment and readable storage medium based on similarity. Background technique [0002] With the rapid development of society and financial industry, credit business is active in the market. The assessment capabilities of the risk control assessment models used by credit institutions are also getting stronger, but the credit fraud models of credit fraudsters are also evolving. Based on the professionalism of the risk control assessment models, financial fraudsters are becoming more and more organized commit crimes with gangs; [0003] The existing risk control evaluation method will conduct a single verification based on various credit information data provided by loan applicants, such as identity information, mobile phone device information, mailbox number, bank card information, and city of residence. The correlation between the credit...

AI Technical Summary

Problems solved by technology

[0003] The existing risk control evaluation method will conduct a single verification based on various credit information data provided by loan applicants, such as identity information, mobile phone device information, mailbox number, bank card information, and city of residence. The correlation between the credit information data of different loan applicants is often ignored, but the credit information data of organized and ganged financial fraudsters always have cross-similar characteristics, so the existing risk control assessment methods Inability to effectively assess ganged fraudsters, reducing the accuracy of credit assessments

Images