The Study Of Yingkou Port Customer Segmentation Based On Convolution Kernels

As port enterprise the major is loading and unloading. Nowadays, the market becomes more and more competitive. In order to make a better business, the management has to be clear that keeping the customers the business already has and attracting more new customers are two key points. For those port enterprises, there are two top tasks: first, how to segment the customers properly; second, how to offer the services according to their own specialties. Therefore introducing customer relationship management to analyze the data which has been collected within the information system is very important for port.Plenty of research has been done in many fields such as telecommunication and so on. Various methods of segmentation have been worked out according to the various conditions. The clustering is widely used to segment the customers and good results have been got. However, for port enterprises, there is still a black. There are not many proper methods of customer segmentation to meet the needs of port. On the basis of deep investigation and analysis of market environment in Yingkou Port, in order to improve competitiveness and the quality of customer service, customer segmentation is put forward. Then the thesis presents a new method that adapt port customer segmentation.The thesis selects the database that strong relate to port customers, then preprocesses the data include data cleaning, data integration, data transformation etc. According to the character of customer data, thesis reorganizes the customer data into tree structure, introduces convolution kernels from statistic and pattern recognition, constructs convolution tree kernels, measures similarity degree between two customer trees to prepare well for customer segmentation.According to the characteristics of tree structure data, combines convolution kernels with clustering algorithm, kernel k-aggregate clustering algorithm based on convolution kernels is designed. After the preliminary conformation, has obtained a better effect. At last, thesis analyses the characters of every class seriously, then gives the appropriate marketing strategies by McKinsey 4P marketing theory to support CRM.According to the study of port customer segmentation, this thesis combines convolution kernels with clustering algorithm, further enriches the kernel clustering algorithm theory and promotes the information development of port enterprises.