The Research On Mobile Corporation Marketing Strategy Of Customer Retain Existing Based On Improved Cluster

At present,the Mobile Corporation marketing strategy analysis of the main problems existing in the system are:on the one hand,the system design focus only on the analysis function of the realization of a simple statistical reports and other traditional marketing strategy,and not pay attention to the new means of marketing the use of data mining,knowledge discovery and utilization;on the other hand,in the Mobile Corporation member for retaining existing marketing analysis system,the most common is the K-means clustering algorithm,the subjective factors but this algorithm is easy to be influenced by the initial number of clusters,modelers,the level of.Therefore,the data mining related theory and algorithm is introduced into the marketing strategy analysis system,abundant marketing strategy analysis of the function of the system,the Mobile Corporation marketing offers more effective marketing strategy analysis tools.Based on the analysis of marketing strategy of A Mobile Corporation involved in the system membership for retaining existing mobile membership marketing,price satisfaction management,maintain high value membership maintenance in three main marketing business,according to the needs of actual work in improving the existing data mining algorithms,to improve the quality of data mining system.The record KBPC mean clustering algorithm for marketing events for quantitative analysis and formula recommended.Most existing algorithms only pay attention to the interest of member of marketing products,that is to say the membership interest high marketing products,the existing algorithms are recommended.But this recommendation algorithm is easy to overlook the Mobile Corporation to the choice of development strategy of marketing products.The paper thinks,interested members to a marketing products for membership have short-term stock marketing strategy of the importance is self-evident,but the long-term value of a marketing products in Mobile Corporation for the member retains the stock marketing strategy is indispensable,in reality,the two are not consistent.Therefore,this paper puts forward a K mean clustering algorithm based on permutation and combination(KBPC,K-means base on Permutation and combination).Finally,the paper respectively in the member for retaining existing mobile membership marketing,price satisfaction management,maintain high value customer relationship in three aspects of the module design requirement analysis,module design and results show.The internal rules of deep mining of marketing strategy,to better guide the A Mobile Corporation stock keep making and marketing strategy is very important.The test results show that,the proposed algorithm and the scheme is feasible,has achieved the anticipated effect.