Design And Implementation Of Financial Agent System Based On Microservice

With the rapid development of global economy and Internet technology,the innovation in the financial field also keeps pace with the times.The way of online investment and wealth management is increasingly favored by the public.Providing users with a good online investment and wealth management environment through highperformance wealth management product sales platform has become a new development direction and demand of securities dealers,banks and third-party platforms.In recent years,with the continuous improvement of China’s Internet financial market environment,the central bank,CSRC and other regulatory institutions have successively issued a series of regulatory policies.The existing financial consignment system can not quickly respond to the new needs brought by compliance transformation.In addition,it did not provide user transaction data analysis and corresponding valuable financial suggestions for specific groups at an appropriate time point.In view of the above problems,in order to improve the user experience and platform competitiveness,this paper designs and implements the financial consignment system.The system adopts an elastic and scalable micro service architecture,which can quickly respond to the changes of system requirements;The clustering algorithm based on LOF algorithm is improved to analyze user data and provide financial position suggestions.This paper expounds the four modules completed by the author: high-end authentication module,transaction module,batch processing module and position module.The high-end authentication module provides users with the subscription qualification of high-end products by means of qualified investor authentication,double record authentication and contract signing;The transaction module provides the process realization and transaction details query related to product purchase and product redemption;The batch processing module provides the ability of offline batch processing of various tasks,and is responsible for the undertaking and conversion of data and the connection between the system and institutions;The position module provides queries on product,position,income and other dimensions.It also supports the summary and analysis of users’ position,income and transaction behavior,gives users the classification level of corresponding characteristics,and provides relevant analysis and suggestions.In terms of technology,the system is based on the spring cloud Netflix distributed service governance framework,uses the spring boot framework to build micro services and realize business code,and reduces the coupling degree and operation efficiency of the system through the basic components and Middleware of various platforms.In order to provide position trading suggestions to users,the system uses the clustering algorithm based on LOF algorithm to improve and analyze the individual position and trading behavior record data of investment users,extract their index characteristics,screen outliers through LOF algorithm,reduce the impact of noise on cluster center calculation,optimize the selection of initial cluster center by using LOF outlier factor to improve the clustering effect,and finally discretize the investor index through K-means algorithm,So as to classify users and rate the position characteristics of users,and then provide users with financial position suggestions.At present,the system has been put into operation online with stable operation,which improves the user experience and platform competitiveness,brings economic benefits to the company and saves a lot of operating costs.