Application Research In Crm Of Bank Based On The Improved K-means Cluster Algorithm

Today, competition in all walks of life are quite intense, especially in the financial sector, competition has become the industry rules, the performance of the grasp, in fact, grasp of the customer, manage client needs more time and more to meet market demand, the more become the industry leader in the competition. Largest sector enterprises to establish the current customer management system, improvement in the level of information to China's financial sector has a large number of customers of electronic data resources, and customer data to improve the understanding and use of service level has an important role. Accurate customer segmentation enables companies to effectively implement customer relationship management strategy. Now the company has experienced a data set collected during the effective use of existing information on how to dig out valuable information to help decision-makers. Traditional technical support has been unable to meet user requirements, data mining technology is a massive data processing, and never completely, noisy and ambiguous data in which to extract hidden information. inductive description of the basis of existing information, by inference prediction for decision-makers to provide meaningful decision support. Application of data mining technology can be divided into meaningful market and customer groups, to help businesses better plan activities and to design new marketing campaigns, through access to customer categories to analyze and predict customer consumption patterns. Therefore, the study of data mining, customer classification, according to the classification results to develop a CRM strategy has important theoretical value and practical significance. This paper focuses on data mining in the financial industry to achieve customer relationship management applications. Paper first reviews the data mining and customer relationship management basic theory. With financial services, analyzes the financial industry customer relationship management capabilities and system structure, while the article on data mining based on clustering analysis, the main research in the financial industry customer relationship management application clustering approach for mining model of customer segmentation and algorithm design and the data mining customer relationship management in the financial industry in the typical application. The main work for the data mining process, the paper through the financial business dataanalysis, customer relationship management enterprise datawarehouse established. K-means algorithm by analyzing the advantages and disadvantages, the paper proposed an improved K-means algorithm. Finally, the paper company to a sales department, for example, the actual work in the data mining process, from goal definition, data preparation, datawarehouse, algorithm and application implementation and results output, select the sales department data for 2009-2010 in order to data mining to achieve, and the corresponding marketing strategy.