The Study, Based On Securities And Customer Clustering Of Mixed Data

In today's competitive business environment, customers have become the company's most important assets, he who knows his customers will control the market. Retaining customers, attracting customers and fully mining the payoff profits of customers are the key to increase the core competitive power for enterprise. This is the same in securities industry. How to have a better understanding of the customers'characteristics and demands, how to design effective products and services and how to increase the customers'income and satisfaction are the key points of customer relationship management (CRM) in the securities industry. The strategy of customer segmentation is the best base for getting customers.Data mining technology is a new business information processing technology, based on customers'internal and external attributes and their consuming behavior characteristics, it can help the enterprise to classify the customers into different groups so as to provide the specific products and services for dissimilar customers. Clustering analysis is the core technology and a very active research direction in the field of Data Mining. It divides the data points into several parts according to the attributes of different objects, making data objects in the same part have a high comparability and those in different parts are as dissimilar as possible.Basing on the expatiation upon the customer segmentation theory and clustering analysis theory, this article makes some improvement on the traditional K-means arithmetic about distance calculation and the choosing of initial clustering center spot, then by making use of the stock customer data, it does a positive research, which leads to a satisfying classification result, and it also shortens the calculation time to a large degree. In the end of the thesis, it offers different marketing suggestions for dissimilar stock customer.