| With the booming development of China's financial industry and the growth of commercial organizations,the data that the banking industry faced has also increased.The demands and concepts of foreign commercial banks were constantly drive the transformation and innovation of Chinese banks in data processing methods.The bank data processing has entered a new era of advanced technologies such as data integration,data warehouse as well as data mining.Firstly,the thesis analyzes the technical difficulties that banks have faced in the process of its operations such as excessive data,diverse business types and lack of analytical methods.Based on the relevant researches and practice results at home and abroad and borrowing the successful experiences of the above data processing technology,the paper takes a well-known commercial bank as the example and starts from the actual business,a bank data comprehensive processing platform integrating data integration,data storage,data analysis and data display was designed and implemented.The main work is as follows:1.A set of data integration schema was designed and implemented to provide integration of various heterogeneous data sources for banks.In the process of researching data integration schema,the paper aims at the characteristics of multiple data sources in banking institutions.Each independent data source was equipped with a separate wrapper,and various types of data were extracted into their own wrappers.With the specified conversion algorithm,the data of different formats were converted into a unified XML Schema and generates an XML file.The schema realizes the extraction and conversion of multi-source data from banks,which can effectively solve the problem of heterogeneous data integration in banks.2.Based on the banking business,a set of database warehouses framework was designed and constructed to achieve the bank data storage and classification.In the process of data storage research,a storage framework of data warehouses was adopted in the paper and combining with the different businesses that implemented in the operation of the bank,the data warehouse framework of the system was separated into two parts: the data warehouse and data mart.The data warehouse was constructed according to the banking business using a topic model and was divided into ten categories of topics.Based on the actual application requirements of data,the data mart was designed by adopting business modeling and star schema.The problem of classified storage of bank data was effectively solved by the framework.3.Different data statistics methods and correlation analysis algorithms were used in the paper to achieve multidimensional analysis and correlation analysis of banking services.A multi-dimensional analysis for banking core business,channel business and financial products is realized.Meanwhile,an optimized Apriori algorithm was adopted to realize the correlation analysis between customers and bank financial products.With the above methods,the analysis of banking business operations was effectively achieved and provided perfect decision support for the future development of banking services.4.Based on the above work,a comprehensive processing platform for bank data was designed and implemented.The system provides the functions of unified integration,classified storage,multi-dimensional analysis and correlation analysis for bank data and by using Flex technology,the function of the data display was achieved.Finally,the test tool was used to test the Web page and the system performance,the reliability and stability were verified.With this system,the problems such as unified integration of multi-source data in banks,classification management of various types of business data and analysis of business operations can be solved effectively.And it plays an important role in enhancing business development and providing decision support. |
PDF Full Text Request