基于综合集成研讨厅的半监督客户关键特征选择模型研究

管理评论 ›› 2020, Vol. 32 ›› Issue (7): 236-245.

基于综合集成研讨厅的半监督客户关键特征选择模型研究

谢玲¹, 陈文婷², 曹瀚文³, 肖进⁴

1. 遵义医科大学医学信息工程学院, 遵义 563006;
2. 西南财经大学科研处, 成都 611130;
3. 华为技术有限公司集团财经, 深圳 518000;
4. 四川大学商学院, 成都 610064

收稿日期:2019-08-26 出版日期:2020-07-28 发布日期:2020-08-08
通讯作者: 肖进(通讯作者),四川大学商学院教授,博士生导师,博士
作者简介:谢玲,遵义医科大学医学信息工程学院副教授,博士;陈文婷,西南财经大学科研处科员,硕士;曹瀚文,华为技术有限公司集团财经工程师,硕士。
基金资助:
国家社会科学基金重大项目（18VZL006）；四川省杰出青年基金项目（2020JDJQ0021）；四川省天府万人计划；四川大学杰出青年基金（sksyl201709）；四川大学科技领军人才培育项目；北京市财政课题"新经济支撑北京高质量发展研究"（PXM2020-178216-000001）。

Semi-supervised Key Feature Selection of Customers Based on Hall for Workshop of Meta-synthetic Engineering

Xie Ling¹, Chen Wenting², Cao Hanwen³, Xaio Jin⁴

1. School of Medical Information Engineering, Zunyi Medical University, Zunyi 563006;
2. Office of Academic Research, Southwestern University of Finance and Economics, Chengdu 611130;
3. Huawei Technologies Co., Ltd., Shenzhen 518000;
4. Business School, Sichuan University, Chengdu 610064

Received:2019-08-26 Online:2020-07-28 Published:2020-08-08

摘要/Abstract

摘要： 客户分类一直是企业客户关系管理（CRM）中最重要的问题之一，而选择出客户的关键特征更是其中的重中之重。在大数据时代，客户数据类别分布不平衡、高维以及大量的无类别标签样本等特征让这一问题变得更为复杂，成为一个复杂的系统性决策问题。为解决这一问题，本文提出基于综合集成研讨厅的半监督客户关键特征选择模型（semi-supervised key feature selection of customers based on hall for workshop of meta-synthetic engineering，SFS-HWME）。该模型邀请5位相关领域的专家确定研究难点并通过定性分析寻找备选方案，然后通过综合集成得到整体解决方案，进一步进行定量分析建模。在定量分析模型中，使用半监督学习（semi-supervised learning，SSL）技术，首先使用初始有类别标签的数据集L训练Adaboost集成模型来预测无类别标签数据集U中样本的类别；接着，使用自组织映射（self-organization map，SOM）算法对数据集U进行聚类并对其中的样本进行选择性标记；然后将这些样本连同标记的类别标签一起添加到数据集L中；最后，使用重抽样技术平衡新的训练集L的类别分布，再训练数据分组处理（group method of data handling，GMDH）深度学习网络选择最优特征子集，并邀请专家从特征子集中选出最合理的。在4个客户分类数据集上进行实证分析，结果表明，和已有的一些模型相比，本文提出的SFS-HWME模型具有更好的关键特征选择性能。

关键词: 综合集成研讨厅, 客户分类, 特征选择, 半监督学习, GMDH, 重抽样

Abstract: Customer classification has always been one of the most important issues in customer relationship management (CRM). Therefore, it is very important to select key features of customers. In the era of big data, unbalanced classification distribution, high dimension, and a large number of samples without label have made this more complex and become a complex systemic decision issue. In order to address this issue, this study proposes the semi-supervised key feature selection model of customers based on hall for workshop of meta-synthetic engineering (SFS-HWME). The model invites five experts in related fields to identify research difficulties, find alternatives through qualitative analysis, obtain a total solution through comprehensive integration and get a quantitative analysis model. The quantitative analysis model uses semi-supervised learning (SSL). Firstly, it uses the data set L with category tags to train the Adaboost integration model to predict the categories of samples in the data set U with unclassified tags; secondly, the data set U is clustered by the self-organization map (SOM) algorithm and the samples are selectively tagged; thirdly, these samples are added to the data set L along with the tagged category tags; finally, the re-sampling technique is used to balance the class distribution of the new training set L, and the group method of data handling (GMDH) deep learning network is trained to pick out the optimal feature subset. The research invites 5 experts to select the most reasonable features. The empirical analysis on four customer classification data sets shows that the proposed SFS-HWME model has better key feature selection performance than some existing models.

Key words: hall for workshop of meta-synthetic engineering, customer classification, feature selection, semi-supervised learning, GMDH, re-sampling

谢玲, 陈文婷, 曹瀚文, 肖进. 基于综合集成研讨厅的半监督客户关键特征选择模型研究[J]. 管理评论, 2020, 32(7): 236-245.

Xie Ling, Chen Wenting, Cao Hanwen, Xaio Jin. Semi-supervised Key Feature Selection of Customers Based on Hall for Workshop of Meta-synthetic Engineering[J]. Management Review, 2020, 32(7): 236-245.

参考文献

[1] 贺昌政,孔力,朱兵,等. 信用卡取现影响因素研究[J]. 管理评论, 2014,26(10):38-45
[2] 张婷婷,贺昌政,肖进. 基于动态分类器集成选择的不完整数据客户分类方法实证研究[J]. 管理评论, 2012,24(6):83-87
[3] Mo L., Xie L., Jiang X., et al. GMDH-based Hybrid Model for Container Throughput Forecasting:Selective Combination Forecasting in Nonlinear Subseries[J]. Applied Soft Computing, 2018,62(1):478-490
[4] Xiao J., Xie L., He C., et al. Dynamic Classifier Ensemble Model for Customer Classification with Imbalanced Class Distribution[J]. Expert Systems with Applications, 2012,39(3):3668-3675
[5] Zhang Y., Zhang Z., Qin J., et al. Semi-supervised Local Multi-manifold Isomap by Linear Embedding for Feature Extraction[J]. Pattern Recognition, 2018,76(4):662-678
[6] Liu J., Wang D., Gao Y., et al. A Joint-L2, 1-norm-constraint-based Semi-supervised Feature Extraction for RNA-Seq Data Analysis[J]. Neurocomputing, 2017,228(3):263-269
[7] 张自然. 寿险公司的财务评估——基于主成分分析的RIDIT方法[J]. 管理世界, 2016,32(3):182-183
[8] Kim K. An Improved Semi-supervised Dimensionality Reduction Using Feature Weighting:Application to Sentiment Analysis[J]. Expert Systems with Applications, 2018,109(5):49-65
[9] 赵鉴华,木克热木·米力克,徐艳梅. 我国各地区FDI引进效率动态分析——基于DEA模型的Malmquist指数方法[J]. 管理评论, 2013,25(4):32-37
[10] 王伟,王洪伟,孟园. 协同过滤推荐算法研究:考虑在线评论情感倾向[J]. 系统工程理论与实践, 2014,34(12):3238-3249
[11] 徐艳艳,黄炜,曾永年. 基于自适应遗传算法的超平面分类及遥感应用[J]. 系统工程理论与实践, 2017,37(3):773-781
[12] Xiao J., Li Y., Xie L., et al. A Hybrid Model Based on Selective Ensemble for Energy Consumption Forecasting in China[J]. Energy, 2018,159(6):534-546
[13] Zhang D., Chen S., Zhou Z. H., et al. Constraint Score:A New Filter Method for Feature Selection with Pairwise Constraints[J]. Pattern Recognition, 2008,41(5):1440-1451
[14] Ang J. C., Mirzal A., Haron H., et al. Supervised, Unsupervised, and Semi-supervised Feature Selection:A Review on Gene Selection[J]. IEEE/ACM Transactions on Computational Biology and Bioinformatics, 2015,13(5):971-989
[15] Ren J. T., Qiu Z. Y., Fan W., et al. Forward Semi-supervised Feature Selection[C]. Pacific-Asia Conference on Knowledge Discovery and Data Mining, Springer, Berlin, Heidelberg, 2008
[16] 严菲,王晓栋. 基于局部判别约束的半监督特征选择方法[J]. 模式识别与人工智能, 2017,30(1):89-95
[17] 戴汝为,操龙兵. 综合集成研讨厅的研制[J]. 管理科学学报, 2002,(3):10-16
[18] 杨涛,张良春. 基于Adaboost集成RBF神经网络的高速公路事件检测[J]. 计算机工程与应用, 2008,44(32):223-225
[19] 肖进,刘敦虎,贺昌政. 基于GMDH的"一步式"客户流失预测集成建模[J]. 系统工程理论与实践, 2012,32(4):807-814
[20] Ivakhnenko A. G., Ivakhnenko G. A., Savchenko E. A., et al. Problems of Further Development of GMDH Algorithms:Part 2[J]. Pattern Recognition and Image Analysis C/C of Raspoznavaniye Obrazov I Analiz Izobrazhenii, 2002,12(1):6-18
[21] 陈宇新,张喆,杨涵方. 基于中国数据的国际管理研究学术进展——基于UTD-24和FT-45的可视化分析[J]. 管理科学学报, 2018,21(11):61-75
[22] Polat K., Güneş S. Detection of ECG Arrhythmia Using a Differential Expert System Approach Based on Principal Component Analysis and Least Square Support Vector Machine[J]. Applied Mathematics and Computation, 2007,186(1):898-906
[23] Zhang N. L., Yi W., Tao C. Discovery of Latent Structures:Experience with the Coil Challenge 2000 Data Set[J]. Journal of Systems Science and Complexity, 2008,21(2):172-183
[24] 蔡淑琴,秦志勇,蒋士淼,等. 基于回归树与衰减函数的在线负面口碑影响力预测[J]. 系统工程理论与实践, 2016,36(6):1496-1504
[25] Harris T. Credit Scoring Using the Clustered Support Vector Machine[J]. Expert Systems with Applications, 2015,42(2):741-750
[26] Niklis D., Doumpos M., Zopounidis C. Combining Market and Accounting-based Models for Credit Scoring Using a Classification Scheme Based on Support Vector Machines[J]. Applied Mathematics and Computation, 2014,234(5):69-81
[27] 邵志国,韩传峰,孟令鹏,等. 基于Logistic的区域交通基础设施生态系统演化模型[J]. 系统工程理论与实践, 2018,38(11):2918-2928
[28] 肖进,刘敦虎,顾新,等. 银行客户信用评估动态分类器集成选择模型[J]. 管理科学学报, 2015,18(3):114-126