The Design And Realization Of Data Mining Technology Of Bank Card

With the world economy growing and the rapid development of information technology, global bank card continued growth, become the main media paid field. In China, the bank is still at the beginning of the sunrise industry, but undeniable, it has become an integral part of life. At present, most bank of our country is still in the horse race circle stage, they blindly expand, and the utilization rate of the card is not high. New products develop too fast, but it is lack of researching consumer behavior. In the future, most business of bank card will face a series of a long-term profitability problem. So, it is necessary and urgent to manage all kinds of data. Because the factors of commercial bank credit risk is various, the traditional credit method cannot adapt to this complex demand or cannot predict credit conditions, but the data mining technology can remedy the insufficiency. Using data mining can in-depth analysis customer’s influencing factors; it can analysis the combinatorial problem of credit risk; it can display data relationship from multi-level and multi-angle; Data mining tool can comprehensively display data relationship and analysis results. In this paper, we go deep into the field of credit card and study data mining technology. We illuminate the application of data mining in data analysis of credit card. The majority of our work is summarized here:1、We analysis many aspects of credit card application, such as credit scoring, fraud detection, customer relationship management and etc. We explain some background knowledge, traditional analysis methods and models, and the new application of data mining technology in dealing with these problems.2、We realized some algorithms, include classify, clustering and outlier detection algorithm. And we applied the algorithms to data analysis of credit card.3、For classify algorithm, we described many algorithms in detail, and bring forward two analysis models of credit card:credit scoring model and customer analysis model.4、For clustering algorithm, we described the algorithm in detail, and bring forward an analysis model of customer consumes behavior.5、In this paper, we go deep into study outlier detection algorithm specially, and realized the algorithm. Then we bring forward an analysis model of customer abnormal behavior of consumption.6、We accomplished a data mining credit card analysis system. In this paper, we describe the architecture, data models and every functional module of the system in detail.