Design And Implementation Of Bank Project Database Information System Based On Data Mining

In the current information age,with the rapid development of data mining and analysis technology,various industries are gradually establishing information databases.There are many disadvantages in traditional bank management,such as inefficiency,time-consuming,difficulty in accessing project data and ineffective using of project historical data for auxiliary decisionmaking.Therefore,it is necessary to establish a project information database and analyze the data in it.In this way,the standardization and scientific level of project management can be further improved.According to the requirements of the project library information system,the system is divided into modules of homepage,project library management,project management,statistical analysis,and setting.Redis is used to cache hot data in the homepage module;a project prediction model is built in the project library management module to assist the financial department in making scientific decisions;the MD5 message summary algorithm is used in the project management module to solve the problem of repeated file uploads;Fine Report is used in the statistical analysis module to perform statistics and report generation;BCrypt algorithm is used to encrypt the password in the setting module.In addition,Spring Security and distributed locks are also used to build a project library information system.In this paper,a decision tree algorithm is used to construct a project prediction model to realize plan management functions and assist the financial department to make scientific decisions.First,collect project information and pre-process the collected data,including data cleaning and feature extraction;and evaluate the performance of the project prediction model constructed using ID3,C4.5 and CART decision tree algorithms through experiments,and CART algorithm is chosen after comparative analysis.Secondly,using the integrated learning AdaBoost algorithm to optimize the CART algorithm,the accuracy of the improved model can reach 89.45%,and the improved algorithm is expressed using AdaBoost-CART.Then,the simulated annealing algorithm is used to solve the problem that the CART algorithm only considers the local optimum in the process of constructing the decision tree.The accuracy of the improved model can reach 91.79%.This model is expressed by SA-AdaBoost-CART.Finally,the SA-AdaBoost-CART algorithm is used to build a project prediction model.The project database was constructed and launched on the Nanjing Branch of the People's Bank of China after the design and implementation of the project library information system,which was affirmed by the users.Therefore,the research results have certain significance for the bank's internal project management.