K-means与朴素贝叶斯在商务智能中的应用  

Application Research of K-Means Clustering and Naive Bayesian Algorithm in Business Intelligence

作　　者：赵敏[1] 倪志伟[1] 刘斌[1] 

机构地区：[1]合肥工业大学,安徽合肥230009

出　　处：《计算机技术与发展》2010年第4期179-182,共4页Computer Technology and Development

基　　金：国家高技术研究发展计划(863)项目(2007AA04Z116);国家自然科学基金项目(70871033)

摘　　要：不同的客户给企业带来的效益并不相同,为了提高企业的客户关系管理水平,采用基于K-means的聚类的Naive Bayesian算法来预测客户价值,从而使企业可以针对不同的客户采用不同的营销策略,为企业决策提供依据。朴素贝叶斯分类模型是一种简单有效的分类方法,它理论基础好,分类精度高,由于朴素贝叶斯分类中的独立假设前提,使得在特征选择步骤能否准确有效的分类显得尤为重要。实验结果表明,该算法能在保证一定的准确率的同时,可以预测出更多的潜在高价值客户。Different customers benefits to enterprise are not the same,in order to improve the level of the enterprise customer relationship management, use the Naive Bayesian algorithm based on the K - means clustering to forecast the customer value ,so that enterprises can use different marketing strategies for different customers. And this will provide a basis for business decisions. Naive Bayesian Classification Model is a simple but efficient solution, and it has solid theory foundation and high accuracy rate of dassifieation, 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.

关 键 词：客户关系管理 RFM模型 朴素贝叶斯 聚类 

分 类 号：TP39[自动化与计算机技术—计算机应用技术]