Bank customer relationship circle analysis method and system

Abstract

The present invention relates to the technical field of financial services, and in particular to a method and system for analyzing bank customer relationship circles. The specific steps of the analysis method are: collecting and organizing the customer's capital flow information in the bank to form capital flow data; A customer is represented by a node, and the relationship formed between customers and customers is represented by an undirected edge, forming a customer fund relationship network. Based on the community division theory gained by Modularity, a divided relationship circle is obtained; in each relationship obtain the importance index of customers in the circle; after obtaining the importance index of all customers in each relationship circle, select the most important customer in the relationship circle as the central customer, and mark its upstream and downstream customers according to its capital flow, Form a customer chain, research and analyze the funds and financial relationship of customers in the customer chain, and then formulate customer marketing plans and risk prevention and control plans. The method has the advantages of many analysis reference indexes and high accuracy of analysis results.

Description

technical field [0001] The invention relates to the technical field of financial services, in particular to a method and system for analyzing bank customer relationship circles. Background technique [0002] With the full realization of interest rate liberalization reform and the accelerated development of direct financing, domestic banking competition has become increasingly fierce, coupled with the slowdown in domestic economic growth, the participation of non-bank financial institutions such as Internet finance companies and non-financial institutions has made It is more difficult for banks to obtain a stable customer base. For banks, if they want to obtain a stable customer base, if they can analyze the status of customers in the industry chain and their relationship with upstream and downstream customers through customer transfer data, and retain data such as customers and funds in the entire industry chain, they will These data can be used to obtain indicators such as...

AI Technical Summary

Problems solved by technology

[0004] Although the above method can realize the analysis of customer transfer relationship, it still has some defects. For example, in the method of community division, only the node with the most occurrences of neighbor nodes is selected for merging, and it is terminated after reaching the number of iterations. When evaluating customer relationships in the community, only the number of transfers is considered as an influencing factor, and factors such as transfer time and transfer amount are not analyzed; in terms of community division, only nodes are considered. If the number of iterations is not limited, Some communities will be very large, while others will be very small, resulting in a large loss of overall information in the sample.

Images