The Customer Segmentation Of Port Based On The Multi-instance Kernel Clustering Algorithm

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.Every efficient customer relationship management bases on the customer segmentation. It is also the most important part of the whole management chain. Plenty of research has been done in those fields such as telecommunication and so on. Variable methods of segmentation have been worked out according to the variable 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 other hand, the data of port business has specialties such as lot of vacancies value, big noise, different organizational method from other businesses, much less customers than retail trade, less descriptive dimensions fro the customer, irregular distribution and etc.The analyses of the port data show us that the traditional data lay-out and the exited clustering algorithms could not be used in the port customer segmentation, so this thesis presents a new three-level data bag by combined with the way in which the multi-instance learning treat the data. Then a multi-instance kernel function is constructed according to the new bag. The partition coefficient and average fuzzy entropy are calculated to decide the best cluster number of the clustering algorithm. Finally the kernel k-aggregate clustering algorithm using the multi-instance kernel is applied to the customer segmentation and gets a good clustering result which provides the managers guidance and evidence of different marketing strategies for corresponding subdivided markets.