Design And Implementation Of Bank Customer Data Mining System

With all of the financial markets open era, China's financial industry, more and more intense competition, a variety of customer information resources, economic value and social value are more important. At present, the rapid occupation of the financial market in a variety of financial products, bank cards due to its portability, and flexibility of Communications, the first to become a leader, and its huge load of information but also the resources to become a veritable center. In addition, the breadth of its transaction data and no regularity allows banks to use the common method in which they need to find valuable information and knowledge is very difficult. Data mining technology makes this work possible, and its task is to get hidden from large databases in which knowledge and information. In general, as the bank card database, which contains massive and complex transaction data; mining method often used in general is one-sided, not completely trustworthy. Therefore, the use of effective data mining strategy is very important.This article comes from Bank of Communications Weifang Branch customer data mining system projects, bank customers through data mining research significance and describes the situation of domestic and foreign research, more in-depth analysis of the bank customer data mining a key problem. Then, discusses the bank customer database mining algorithms and critical thinking problems, but also discusses the application of traditional data mining techniques to the bank card data mining steps, and this several steps involved in the methods are described, including:bank customers how to design and implement the system of data mining system; bank customer database mining system design architecture; bank customer database mining system implementation techniques, which also includes parallel data preprocessing, transaction-based online analytical processing Mining design features and implementation. Finally, it is the bank customer relationship management and customer data mining system and results of association rules. This article addresses the key issues are the following:First, Bank customer data mining system is a data cleansing, the transaction characteristics of the mining, trade association mining, clustering and trend analysis of customer features such as composition. Data cleaning is a distinction between the type of transaction under the transaction code, and the results stored in the database. Trading Characteristics of excavation is the area where according to the customer, branch, card number and other characteristics, classification of types of transactions.Transaction association mining is to find customer support and reliability of specified related party transactions. Customer clustering is to have some common characteristics of customers are classified. Trend analysis is to use recent transaction data to predict the future short-term transaction amount.Second. Data mining system by the system control module, data preprocessing module, data mining analysis module and report output modules and other components. System Control Module task is to accept the customer call to the appropriate operating instructions to complete the request sub-module;data preprocessing module is responsible for data cleansing and import; data mining analysis module responsible for the characteristics of the target data mining, association analysis, cluster analysis and trends prediction; report output module put the result of data mining analysis of the intuitive way to show it.