Abnormal trading behavior monitoring method, device, computer device and storage medium

Abstract

The embodiment of the invention discloses a method, a device, a computer device and a storage medium for monitoring abnormal trading behavior, wherein, the method comprises the following steps of: acquiring trading behavior data in a trading system in real time; If the abnormal transaction behavior monitoring request is received, determining the abnormal transaction behavior type to be monitored according to the abnormal transaction behavior monitoring request; Calling a preset monitoring rule matching with the abnormal transaction behavior type to be monitored from a preset database; Calculating an evaluation index when the real-time acquired transaction behavior data is an abnormal transaction behavior type to be monitored according to a preset evaluation exponential function associatedwith the called preset monitoring rule and monitoring parameters and parameter thresholds; When the evaluation index is greater than or equal to a preset alarm threshold, generating corresponding alarm prompt information. The invention can quantify the abnormality degree of the transaction behavior, so that the supervisor continuously monitors all the transaction behavior of the customer, and classifies and handles the transaction behavior according to the abnormality degree.

Description

technical field [0001] The invention relates to the field of data processing, in particular to a method, device, computer equipment and storage medium for monitoring abnormal transaction behavior. Background technique [0002] The securities and futures industry is a highly digital and intelligence-intensive industry, and the game between investors and regulatory agencies is full of games in the securities trading process. In order to ensure the health and order of the capital market, the law requires relevant institutions to supervise the abnormal trading behavior of investors. After nearly 30 years of development in the securities market, my country has established a relatively complete system of laws and regulations for the supervision of abnormal trading behavior. As a securities business institution that directly faces investors, securities companies are also responsible for monitoring abnormal transactions. Securities companies urgently need to enhance their own monit...

AI Technical Summary

Problems solved by technology

[0004] 1. Unable to evaluate pre-alarm trading behavior

If the parameters of the customer's transaction behavior do not exceed the threshold, the system generally does not prompt, so it cannot evaluate and calculate each transaction order of the customer, and it is not easy to find potential problems and give timely response

[0005] 2. The early warning mechanism is complex and the threshold is discontinuous

The parameter management of this method is complicated. The more levels of threshold setting, although the management is finer, the more difficult it is to manage.

There are jumps in the threshold setting of adjacent levels, and the number of combinations of customer transaction order parameters is infinite, so it is impossible to continuously evaluate the abnormality of customer transaction behavior

[0006] 3. The degree of abnormal transaction behavior is not convenient for comparison and ranking

In the algorithm for identifying abnormal behavior types, generally speaking, the larger the parameter value, the higher the degree of abnormality, but the same algorithm often involves multiple parameters, and there are multiple algorithms and multiple parameter combinations for the same type of abnormal behavior. Different abnormal behavior parameters different, which makes it very difficult to use the parameter threshold method to compare the degree of anomaly between transaction behaviors

[0007] 4. It is impossible to intuitively and finely express the abnormal degree of transaction behavior

After the parameter threshold method identifies abnormal trading behavior, its abnormal degree can generally only be divided into two categories of "early warning" or "alarm" for reminders, which cannot achieve refined management and intuitively express the abnormal degree of trading behavior. Meet the requirements for subsequent classification and disposal

Images