Design And Implementation Of Commercial Bank Customer Relationship Management System Based On Decision Tree Algorithm

With the economic development and social progress,customer data has become the key in the competition of commercial Banks.Presently,domestic commercial Banks can’t provide specific products and services to customers.The main reason is the lack of effective strategies to evaluate the value of individual consumers and standardized methods to analyze customer data.Based on the method of data mining,this paper initially provides the system design scheme of bank customer relationship management.Based on the classical ID3 algorithm,the WID3-NB algorithm is implemented by using naive bayes and balance function design.The prediction classification model constructed by this algorithm is to correct the desired information amount of subtuples by using balance function,so as to overcome the multi-value bias problem of the classical ID3 algorithm.At the same time,the number of layers in the ID3 tree is limited,and the naive bayes classifier is used at the leaf nodes to avoid the problem when the number of data attributes is large,the number of recursion increases,leading to the low efficiency of the algorithm.Through experimental analysis,the accuracy and efficiency of WID3-NB algorithm in classification prediction are better than the classical ID3 algorithm.Finally,the WID3-NB algorithm is applied to the customer relationship management system of commercial Banks to provide support for decision-making and play an important role in reasonably allocating limited resources and improving the competitiveness of commercial Banks.the traditional practical SSH framework and classification prediction technology are used to design and implement a commercial bank customer relationship management system including customer information management,decision analysis management,marketing management and other functions.The system front end pages using jquery and Jfreecharts technology,the back-end using the traditional SSH framework,database using MySQL,also uses WID3-NB classification prediction algorithm to excavate the implicit information between commercial bank customer information deeply.The decision tree classification model is generated to obtain the rules of prediction classification,provide objective and right decision support for bank decision makers,allocate limited resources reasonably,and develop personalized services for VIP customers to improve the profits of commercial Banks.