E-bank account fraudulent conduct and risk detecting method and system

Abstract

The invention discloses an E-bank account fraudulent conduct and risk detecting method and system. The method includes model construction and risk analysis. A mainstream big data processing platform is utilized for extracting an overall index and individual indexes from original data and analyzing the indexes. Risk evaluation models for different accounts are constructed and dynamic adjustment and update of the models are performed according to newly input data. When a to-be-detected-account is being evaluated, corresponding indexes of the account are extracted and risk grades of the indexes are subjected to weighing accumulation, so that the risk evaluation value of the account is obtained. Further, the to-be-evaluated account is graded based on a risk grade model obtained through analysis of distribution of all history risk evaluation values. Finally, accounts with comparatively high risk grades are obtained. According to the invention, by utilizing a method combining the overall index with the individual indexes, features are found in data and different features are employed for different accounts. Therefore, fraudulent conduct detection error rate is reduced. At the same time, data processing efficiency is improved.

Description

technical field [0001] The present invention relates to the field of electronic commerce, and more specifically, to a method and system for detecting fraudulent behavior and risks of electronic bank accounts. Background technique [0002] Electronic banking is an extension of the traditional counter business of banks, which can make banking business no longer restricted by business location and business hours, and provide people with the financial services they need at any time. While it brings convenience, it also faces many security risks, especially as fraud cases continue to increase, it will bring huge losses to banks and their users. At present, in order to better prevent the occurrence of e-bank fraud, the method of generating rules based on expert systems is mainly used to detect fraud. However, the rule-based method usually brings problems such as rule update lag, high false positive rate, and low matching efficiency. , so that the interests of banks and users cann...

AI Technical Summary

Problems solved by technology

However, there is no detailed description of the key rule generation process in the entire system, resulting in the lack of strong support for the subsequent generation of risk scores and the process of risk assessment based on risk thresholds

Moreover, as mentioned above, the rule-based detection method is no longer able to adapt to the reality faced by the current e-bank: firstly, this method cannot identify new risks; secondly, when the number of rules is too large, the analysis and detection process will affect the system Processing power brings obvious pressure

Images