THE EMPIRICAL STUDY OF APPLYING NEURO TO ESCALATE MARKETING PERFORMANCE

摘要：

We develop a F-measure performance evaluation system by combining with RFM criteria and Bayesian classification, Decision Tree Induction, Gini Index, Neuro and by analyzing the factors with recall related and precision.By using this system, more profitable customers can be discovered and less unprofitable customers will be missed.The research collects and analyzes references on promoting buying rate; moreover, it introduces R-F-M (recency, frequency, monetary amount) criteria, brings up the idea of identify each individual customer to promote both the marketing profit and the customers lifetime value.The result shows that marketing performance derive from Neuro-weighted RFM model has the advantage over traditional RFM model by 27.80％.

关键词： Data mining Rfm Bayesian classification Decision tree induction Gini Neuro

作者: TONC-SHENG CHEN

作者单位: Department of Computer Science, Xiamen University, Xiamen 361005, China

会议类型: 国际会议

会议名称: 2007 International Conference on Machine Learning and Cybernetics(IEEE第六届机器学习与控制论国际会议)

会议地点: 香港

会议语种:英文

页码: 287-292

在线出版日期: 2007-08-19（万方平台首次上网日期，不代表论文的发表时间）