混合Markov与Bayes的客户欠费预测模型

doi:10.3969/j.issn.1000-1565.2016.05.014

河北大学学报(自然科学版) ›› 2016, Vol. 36 ›› Issue (5): 535-540.DOI: 10.3969/j.issn.1000-1565.2016.05.014

混合Markov与Bayes的客户欠费预测模型

吴舒霞¹,陈炼¹,高胜保²

收稿日期:2016-02-29 出版日期:2016-09-25 发布日期:2016-09-25
作者简介:吴舒霞(1991—),女,江西南昌人,南昌大学在读硕士研究生. E-mail:352901867@qq.com
基金资助:
国家自然科学基金资助项目(61463033);江西省教育厅科学技术研究项目(GJJ14136)

A prediction model of customer arrears based on hybrid Markov and Bayesian

WU Shuxia¹,CHEN Lian¹,GAO Shengbao²

1.Information Engineering College, Nanchang University, Nanchang 330031, China; 2.Jiangxi Branch of China Telecom Co, Nanchang 330029, China

Received:2016-02-29 Online:2016-09-25 Published:2016-09-25

摘要/Abstract

摘要： 为有效规避客户欠逃费和实现差异化处置,针对具有长期定时付费特征的后付费类服务,提出混合马尔科夫与贝叶斯的客户欠费预测模型,基于全客户多要素信息增益分析,给出潜在欠费客户的欠费概率,为客户欠费预警和处置提供全面、客观、精细的决策信息,并支持客户差异化处置.首先,基于其付费特点,建立k序马尔科夫模型,计算客户的初始欠费概率;然后,融合客户基本属性、行为特征和欠费信息等要素,基于条件互信息和爬山法生成目标贝叶斯网络,对初始欠费概率予以修正,形成客户最终欠费概率;最后,基于实际数据进行实证分析,验证了该模型的有效性.

关键词: 后付费客户, 欠费预测模型, 混合马尔科夫, 贝叶斯

Abstract: In order to analyze the post-paid services with the characteristics of long term and on time pay,we put forward a prediction model based on hybrid Markov and Bayesian. It is based on the multi-factor information gain of all the customers,and computes the potential owe customers' probability of arrears.Moreover,it can provide comprehensive,objective and subtle decision information to the warning of customer arrears and disposal,and it can support differentiation treatment.First of all,we build the k-order Markov model based on the characteristics of the pay,then calculate the customers' initial probability.Secondly,we merge the customers' basic attributes,behavior feature and own information.Then,using the conditional mutual information and the hill climbing algorithm to generate the target Bayesian network to modify the initial probability of arrears,which form the final client own probability.Finally,through experiment by using the real data,we prove that this predict model is efficient in customer prediction.

Key words: post-paid customer, probability prediction model, hybrid Markov, Bayesian

中图分类号:

TP181

吴舒霞,陈炼,高胜保. 混合Markov与Bayes的客户欠费预测模型[J]. 河北大学学报(自然科学版), 2016, 36(5): 535-540.

WU Shuxia,CHEN Lian,GAO Shengbao. A prediction model of customer arrears based on hybrid Markov and Bayesian[J]. Journal of Hebei University (Natural Science Edition), 2016, 36(5): 535-540.

参考文献

[1] 李斌杰.住院病人的欠费预测模型研究[J].中国数字医学,2010,26(12):60-62.DOI:10.3969/j.issn.1673-7571.2010.012.019. LI B J.The research of inpatient delinquency forecasting model[J].Health Information Forum.2010,26(12):60-62.DOI: 10.3969/j.issn.1673-7571.2010.012.019.
[2] 王栋,董理,占守义.基于分类技术的电信欠费预测模型[J].计算机工程与应用,2003,38(7):123-125. WANG D,DONG L,ZHAN S Y.Owing fee model in telecommunication based on classifying[J].Computer Engineering and Appliations,2003,38(7):1123-125.
[3] 李学鹏,张国基.基于支持向量机的电信欠费客户分析模型[J].计算机应用,2006,5(12): 214-215.
[4] 李玉华,李栋才,毕威,等.混合马尔科夫预测模型及其在反洗钱中的应用研究[J].计算机科学.2011,38(7):170-174. LI Y H,LI D C,BI W,et al.Hybrid Markov prediction model and research of application in anti-money laundering[J].Computer Science,2011,38(7):170-174.
[5] DERRODE S,PIECZYNSKI W.Unsupervised classification using hidden Markov chain with unknown noise copulas and margins[J].Signal Processing,2016,128:8-17.DOI:10.1016/j.sigpro.2016.03.008.
[6] 郑莉华,陈佳.基于贝叶斯网络的电信话费欺诈的模型研究及应用[J].计算机应用,2008,28(2):510-512. ZHENG L H,CHEN J.Research and application of telecom charge fraudulent model based on Bayesian network[J].Computer Applications,2008,28(2):510-512.
[7] 金焱,胡云安,张瑾,等.互信息和爬山算法相结合的贝叶斯结构学习[J].计算机应用与软件,2012,29(9): 122-125.DOI:10.3969/j.issn.1000-386x.2010.09.032. JIN Y,HU Y A,ZHANG J,et al.Bayesian network structure learning combining mutual information with hill climbing algorithm [J].Computer Applications and Software,2012,29(9): 122-125.DOI:10.3969/j.issn.1000-386x.2010.09.032.
[8] GHEISARI S,MEYBODI M R.BNC-PSO: structure learning of Bayesian networks by particle swarm optimization[J].Information Sciences,2016,348:272-289.DOI:10.1016/j.jns.2016.01.090.
[9] 朱明敏.贝叶斯网络结构学习与推理研究[D].西安:西安电子科技大学,2013. ZHU M M.Research on structural learning and inference in Bayesian networks[D].Xi'an:Xidian University,2013.
[10] JIANG J,WANG J,YU H,et al.Poison identification based on Bayesian network: A novel improvement on K2 algorithm via Markov blanket[J].Lecture Notes in Computer Science,2013:173-182.DOI:10.1007/978-3-642-38715-9_21.

混合Markov与Bayes的客户欠费预测模型

A prediction model of customer arrears based on hybrid Markov and Bayesian

PDF (PC)

可视化

被引次数

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 8

编辑推荐

Metrics

本文评价

[1]	毛华,刘谦,连萌璇,王刚,张植明. 地下工程岩体质量评判的三支决策方法[J]. 河北大学学报(自然科学版), 2023, 43(3): 330-336.
[2]	戚元星,崔双喜. 集成学习在电网假数据入侵检测中的应用[J]. 河北大学学报(自然科学版), 2022, 42(1): 105-112.
[3]	刘振鹏,张庆文,李泽园,刘嘉航,董姝慧,赵永刚. 软件定义网络中基于贝叶斯ARTMAP的DDoS攻击检测模型[J]. 河北大学学报(自然科学版), 2021, 41(6): 728-733.
[4]	刘佩鑫,于洪志,徐涛. 基于朴素贝叶斯的档案分类研究[J]. 河北大学学报(自然科学版), 2018, 38(5): 549-554.
[5]	肖仙谦,朱俊平,景旭,马巧娥. 基于贝叶斯方法的单分类入侵检测技术[J]. 河北大学学报(自然科学版), 2014, 34(1): 7-13.
[6]	霍亮,杨柳,张俊芝. 贝叶斯与k-近邻相结合的文本分类方法[J]. 河北大学学报(自然科学版), 2012, 32(3): 316-319.
[7]	梁宏胜,徐建民,成岳鹏. 一种改进的朴素贝叶斯文本分类方法[J]. 河北大学学报(自然科学版), 2007, 27(3): 327-331.
[8]	吴志峰,田学东. 人名、机构名在基于概念的文本分类中的应用研究[J]. 河北大学学报(自然科学版), 2004, 24(6): 657-661.