0 Preface
With the improvement of people's living standards and the rapid development of e-commerce, people can no longer meet the long queues in the business hall of banks, and start to complete their financial management processes in a quick and convenient way. As a product of the close integration of information technology and banking business, online banking is booming all over the world, with continuous innovation and unlimited potential. How to use the existing online banking customer resources and data resources of other banking systems, and conduct real-time analysis, evaluation and prediction to provide leaders with decision support for business development and improvement, and solve the existing "Eighth-Eighth Effect", which is 20% Of customers provide a comprehensive return of 80%, which is the key to the healthy, rapid and successful development of online banking.
In order to solve the shortcomings of the current online banking, on the basis of the full investigation and analysis of the evaluation system of online banking at home and abroad, the key technologies of the online banking evaluation system based on the Web model are studied. Using SOA (Service-Oriented Architecture) framework and Aajx and other technologies to design and implement an online banking customer evaluation system based on B / S structure.
1 Research on key technologies
1.1 Service-oriented architecture
Web application architecture is also constantly being explored. According to the complexity of data and control in the application, Web applications can be divided into four categories: Brochure Web ApplicaTIon, Service-Oriented Architecture, Data Intensive ApplicaTIon, and information system applications ( InformaTIon System ApplicaTIon).
SOA is a software design method for Web applications. It provides services to end-user programs or other service programs through published or discoverable interfaces. In fact, SOA is a collection of services. These services communicate with each other, either as simple data interaction or as a collaboration of two or more services. Web Services (Web Services) are interfaces that describe a set of operations that can be accessed through standard XML messages. A Web service performs a specific task or set of tasks. Web services are described with a set of standard XML annotations, called service descriptions. It explains all the details needed to interact with the service, including the message format, transmission protocol and location.
1.2 AJAX technology
AJAX (Asynchronous JavaScript and XML) is not a technology, it is a powerful combination of several technologies. AJAX includes: Standard expressions (Standards-based Presentation) using XHTML and CSS; dynamic display and interaction using DOM (Document Object Modem); data exchange and operation using XML and XSLT; asynchronous data transmission using XML-HttpRequest; use JavaScript binds all of these.
AJAX differs from traditional web application HTTP response synchronization methods. AJAX implements an asynchronous response to HTTP requests, which usually generates a user action for HTTP requests. Now the JavaScript calls the AJAX layer instead. The response of any user action no longer requires direct transmission to the server, such as simple data verification, in-memory Data editing, and even some page navigation, the engine can handle it. If the engine needs to obtain data from the server in response to user actions, if the data to be processed is submitted to another interface code, or receives new data, the engine makes these tasks asynchronous, usually using XML without delaying the user interface interaction, improve The response speed of the system.
2 Design of the evaluation system
The system adopts the B / S architecture, and based on the investigation of relevant units and personnel, an online banking customer evaluation system is designed. The system mainly involves the following user roles, and its top-level user diagram is shown in Figure 1. 
(1) System administrator. Responsible for user data maintenance, role assignment, browsing model, etc.
(2) Branch operator. Obtain the bank's (within the jurisdiction of AA) customers' online banking contribution status, transaction type and transaction size forecast, more professional forecast data, and make special case forecasts for special customers (non-ICBC customers).
(3) Sub-branch operator. Obtain the forecast of the contribution of the customers within this sub-branch after the opening of online banking, the type of transaction, and the scale of the transaction, and make a single case forecast for special customers (non-ICBC customers).
3 Implementation of the evaluation system
The system article uses AJAX technology Web pages to implement the client that calls the SOA framework service, which solves the problems of cross-platform and programming language and user-free installation and deployment. Web applications that use AJAX can provide more functions for the presentation layer of SOA, and can directly call Web services using a browser to better respond to changing business needs. Due to space limitations, only the implementation of customer data collection and cleaning, and customer contribution value prediction functions are introduced here.
3.1 Online banking customer data collection and cleaning
Data cleaning (Data Cleansing) function is to detect errors and inconsistencies in the data set, and use manual or automated tools to remove or correct them to improve data quality. The system uses SQLServet 2005 Integration Service (Integration, Service) to achieve automatic, regular completion of data collection, cleaning and other tasks for the data warehouse.
The control flow structure in the package object that processes online banking customer data is shown in Figure 2. 
The control flow first deletes the expired online banking customer data in the data warehouse through the "delete online banking data" task, and then uses the "import data from online banking server" task to complete the data import from the Oracle server to the SQL Server server. Then perform the "generate online banking data" task, the object completes the data cleaning by running a stored procedure. If any task fails, go to the "send e-mail task" object, if successful, go to the "cluster processing" task, using the "basic processing" embedded Visual Basci. The .NET scripting language runs clustering programs located locally. The FCM algorithm is used to cluster the online banking customer data.
3.2 Realization of customer contribution value prediction
The customer contribution prediction function is divided into three sub-functions: "contribution value prediction", "contribution level prediction", and "advanced prediction". In this function, when the user enters the forecast date and account opening bank (the default is all sub-branches), only the top N digits ranked by contribution value from largest to smallest can be displayed. The N value can be specified in this interface. You can also save the model prediction results as an Excel file. The key codes to realize customer contribution value prediction are as follows: 
4 System operation and validity verification
Before the system is put into use, account managers market customers based on their own experience or other forecast data. The limited number of intersections of high-quality customer sets formed based on experience and real online banking is not sufficient to assist account managers in effective marketing.
According to the prediction results of high-end online banking customers in the second quarter after the system is put into production, high-quality customers are mainly concentrated in the main urban area. Through the correlation analysis of the decision tree algorithm, it is found that there are strong correlations between different transactions of online banking. Table 1 shows that “bulk payment†and “enterprise finance room†and “settlement agency†and “bank-enterprise interconnection†are highly correlated. 
Table 2 is the statistical table of online banking customer account opening in the second quarter (the fourth quarter of the year) and the same period last year of the "B / S Mode Online Banking Customer Evaluation System". 
As can be seen from Table 2, the growth rate of customers in the fourth quarter of 2007 and the fourth quarter of 2008 was 66%, but high-end customers increased by 311%; mid-end customers increased by 130%; low-end customers only increased by 34 %. According to previous analysis, the total growth of 66% includes nearly 20% of natural growth each year, 33% growth of the "New Year's Feedback" marketing in the fourth quarter of 2007 (this item is based on previous similar marketing data statistics) and other improvements (Including "online banking system"). Therefore, the system has a small impact on the total customer development; in addition to the above factors in the growth of mid- to high-end customers, the "B / S Mode Customer Evaluation System" has a greater impact on it. According to the analysis of the tracking and evaluation data after the system runs, it shows that the accuracy of the system's target prediction is acceptable. The prediction accuracy of online banking high-end customers is particularly prominent; the online banking customer transaction analysis has also reached the expected target. In the next step of improvement, the following measures can be used to improve the accuracy of bank customer transaction analysis:
(1) Re-analyze and study customer usage patterns in conjunction with business field personnel, and then adjust the corresponding model structure;
(2) Make further adjustments to the association mode and scope of the input attributes of the existing model to make it more accurately describe the relationship with the predicted attributes.
5 Conclusion
Here, the SOA architecture model and AJAX technology are used to implement the online banking customer evaluation system based on the B / S model, and the customer evaluation system that integrates data collection, model building, model evaluation, and high-end customer forecasting is implemented. Through the practical verification of the platform, not only the existing resources are effectively used, but also the high-end customers in the quarter after the system is put into production increased by 10 percentage points compared with the same period of last year, solving the previous "28" effect. Won more customers for the bank and created higher efficiency.
Led Inground Light are using a famous led chip, CREE,OSRAM,Epistat, Bridgelux etc.,energy saving and environmental protection, long service life. Have RGB with DMX control or single color white, warm white, yellow, red, green, blue etc. for option. Die cast aluminum Lamp body, 304 stainless steel casting surface cover, high performance on corrosion resistance. With plastic embedded parts for convenient installation and maintenance. Widely used in School, hotel, parks, path, garden, roads, building square ground floor shopping malls, pedestrian streets, corridor lighting and other outdoor underground decorative places,ect..
Led Inground Light
Led Inground Light,Exterior Led Lighting,Outdoor Ground Lighting,Ground Lights For Driveways
Guangdong guangzhidian lighting Co., Ltd. , https://www.gzdlighting.com