The Research And Design Of A Lightweight J2EE-based Securities Data Analysis System

Along with development of the computer technology and Internet, securities data analysis methods based on information technology have been used by all kinds of bank enterprise and investment institutions. Securities data is essentially a lot of time series data. There is very important significance to forecast the securities market trend for securities investment activities based on the traditional methods of statistical analysis and using the data mining technique.In data mining areas there have been some mature classification prediction method to describe data classification mode or forecast future trends. In this paper based on naive bayes classification method we make the securities market trend forecast. In forecast analysis we will use the method based on calculation of stock index in the traditional securities analysis, also will involve the relevant theory of statistics and make some change for the secirities data analysis.In the practical application of securities analysis method, this paper designs an analysis system based on J2EE lightweight archcitecture. It provides service calls of data analysis method, inquiry function of the analysis result and expert evaluation function by network application platform. We can effective utilization of the bayesian classification method to make securities analysis, and it is very convenient to assess the application result by the system.There are three layouts in the implemtation of lightweight J2EE-based system. The web layout is implemented by JSF framework, the business layout for business logic control and service management is implemented by Spring framework, and the database system access layout is implemented by Hibernate and JDBC.