摘要
在银行零售业务交易系统中,如何在大量客户数据交易网络中挖掘出影响力高,潜在价值高的重要发展客户,从而制定相应的业务营销计划,对银行来说是一件至关重要的事情.本文提出一种基于PageRank的改进算法——IER(Improved Enhanced-RatioRank)算法,该算法以客户作为节点,以主动交易金额构成出链权重因子作为有向边,构成一个客户交易网络有向图,通过添加交易次数活跃因子和时间有效性因子等重要因素,从多维角度可以精准有效地挖掘出重要发展客户.最后,利用RFM(Recency, Frequency, Monetary)模型来验证实验结果.实验结果表明,所提算法在银行零售业务交易系统中挖掘重要发展客户有良好的效果.
In a bank's retail business transaction system,it is a crucial matter for the bank to dig out the important development customers with high influence and potential value in a large number of customer data transaction networks,so as to develop a corresponding business marketing plan.In this paper,we propose an improved algorithm based on PageRank-IER(Improved Enhanced-RatioRank)algorithm,which takes customers as nodes and active transaction amount constitutes out-chain weight factors as directed edges to form a directed graph of customer transaction network,by adding important factors such as active factors of transaction times and time effectiveness factors,important development customers can be accurately and effectively mined from a multi-dimensional perspective,and finally the RFM(Recency,Frequency,Monetary)model is used to verify the experimental results.The experimental results show that the algorithm proposed in this paper has good effect in mining important development customers in retail business transaction system of banks.
作者
王嵘冰
张子扬
柯娜
WANG Rong-bing;ZHANG Zi-yang;KE Na(College of Information,Liaoning University,Shenyang 110036,China)
出处
《辽宁大学学报（自然科学版）》
CAS
2023年第1期20-27,共8页
Journal of Liaoning University：Natural Sciences Edition
基金
辽宁省社会科学规划基金项目(L21BGL026)。
关键词
PAGERANK算法
交易网络有向图
出链权重因子
交易次数活跃因子
时间有效性因子
RFM模型
PageRank algorithm
directed graph of customer transaction network
out-chain weight factors
active factors of transaction times
time effectiveness factors
RFM model
