Clustering And Na Ve Bayesian Algorithm In Customer Value Forecasting

In recent years, marketing has become more competitive, the various businesses excel not only in product quality, but also in business management and marketing of information technology has also made great improvements began to shift from a product-centric to a customer-centric strategy.With the increase in merchandise sales, generate a large number of customers, the sales department, the importance of these customers are different, in practice, due to different levels for different number of customers the business value of different, simultaneously cost of maintaining customer relationships by spending limits, the more important for business customers only activities that predict customer value to put forward new demands. With the development of information technology, data mining used in corporate information technology began to appear, which makes enterprise greatly enhanced data processing capabilities.This Na?ve Bayesian classification using data mining techniques combined with K-means clustering algorithm to study the importance of customer issues to the data warehouse for customer data objects, trying to build on the current data value of the model, and model appropriate analysis, to identify factors that predicted with a variety of valuable information is hidden to help improve enterprise customer relationship management, enterprise development strategies to further provide the basis for reference, So that enterprises can use different marketing strategies for different customers. And this will provide a basis for business decisions. Require enterprises to provide the marketing activities of our customers the greatest value. This is not only to increase the customer value in the number of customers, but also to increase the customer list in the number of high-value customers. In order to carry out more targeted marketing, enterprise value to be relatively high for those customers better service, accurate forecasts an important customer, the right to establish the target market is the enterprise customer relationship management the key. Naive Bayesian Classification Model is a simple but efficient solution, and it has solid theory foundation and high accuracy rate of classification, an effective feature selection is very important for an NB-based classifier which uses the conditional independence assumption. Experimental results show that the algorithm can guarantee a certain degree of accuracy and can predict more high-value potential customers.