Method and device for dividing customer relationship network

Abstract

The invention provides a method and a device for dividing a customer relationship network. The method comprises the following steps: dividing a client node in a customer relationship network into a plurality of customer groups; cyclically calculating modularity of the customer relationship network, and performing customer group mergence processing on the customer relationship network according tothe calculated modularity until the customer relationship network is merged into a customer group; confirming the customer relationship network with the maximum modularity obtained through calculationin the process of cyclically calculating the modularity of the customer relationship network and in the process of carrying out customer group mergence processing on the customer relationship networkaccording to the modularity obtained through calculation, and setting the customer relationship network with the maximum modularity to be a customer relationship network division result. The processing process uses the modularity of the customer relationship network to realize customer group division of the customer relationship network and find out the optimal customer group division result therein, so as to achieve the purpose of optimal division of the customer relationship network and realize quantitative identification of the customer groups.

Description

technical field [0001] The invention relates to the technical field of risk management, in particular to a method and device for dividing customer relationship networks. Background technique [0002] At present, under the background that the domestic economy is in the "three phases" superimposed and the downward pressure is increasing, the economic growth rate has dropped, the economic structure has been adjusted in depth, the business model and customer structure of enterprises have changed, and the forms of customer financial risks are diversified. When business problems and financial crises occur, there will often be a domino effect, and the risk will quickly spread to the entire customer relationship network, causing the entire enterprise in the relationship network to be in trouble, and regional and systemic financial risks are very prone to occur, which continuously increases the difficulty of risk management . [0003] The traditional risk management model still focu...

AI Technical Summary

Problems solved by technology

[0002] At present, under the background that the domestic economy is in the "three phases" superimposed and the downward pressure is increasing, the economic growth rate has dropped, the economic structure has been adjusted in depth, the business model and customer structure of enterprises have changed, and the forms of customer financial risks are diversified. When business problems and financial crises occur, there will often be a domino effect, and the risk will quickly spread to the entire customer relationship network, causing the entire enterprise in the relationship network to be in trouble, and regional and systemic financial risks are very prone to occur, which continuously increases the difficulty of risk management

[0003] The traditional risk management model still focuses on a single customer and focuses on analyzing the financial risk of a single customer, often ignoring the risk correlation between customers

Financial institutions should gradually establish a credit risk management and control system based on customer relationship networks to realize risk management from individual customers to customer relationship networks. However, there is still a lack of quantitative identification methods and methods for customer groups with strong risk conduction and infection characteristics. means

Images