Application Research Of Data Mining In Bank Customer Relationship Management

With the reform and progress of the financial market in the new era and the innovation of Internet technology,customer-centric customer relationship management(CRM)is one of the hotspots of commercial banks.Effective CRM can bring too much profit to the bank.It combines with supply chain management and enterprise resource planning to form the basic framework of enterprise informationization.The core idea of customer relationship management is to use modern information technology to focus customer management as an important strategic resource,and combine the information obtained by the customer relationship management system with the actual business process of the enterprise.Realize the sharing of resources of internal customer information,continuously improve customer value,and ultimately bring excess profits to the company.Customer relationship management based on data mining is the application of data mining in the bank,analyzing and mining customer information to achieve accurate classification of customers.Data mining technology can be used to mine customer consumption patterns and consumption trends,and to predict customer profitability to banks.It can support the decision-making of commercial banks.How to effectively use data mining technology is the key to the customer relationship of commercial banks.Through the analysis of various types of information of customers,predict their credit level.Therefore,this paper uses data mining technology to implement customer relationship management.Special emphasis is placed on the application of data mining in customer classification and credit analysis.After comprehensively comparing Chinese and foreign literatures,this paper applies data mining to customer relationship management from two aspects: customer credit analysis and customer classification.Customer credit analysis trains and models data through decision tree and neural network algorithm,compares two data mining methods,formulates credit analysis rules,and uses this model to predict customer credit,with an accuracy rate of 85.5%.The customer classification uses the C5.0 decision tree algorithm to train the data set,formulate corresponding rules,and use this rule to classify customers.Finally,according to the training model,commercial banks apply for housing loans to customers in five categories.This paper has 15 pictures,6 tables,and 62 references.