基于边界样本欠取样支持向量机的电信用户欠费分类算法

doi:10.11959/j.issn.1000-0801.2017208

Abstract

Abstract:

Telecom users’ arrears forecasting is a classification problem of unbalanced data set.To deal with the problem that the traditional SVM on the unbalanced date set had a low detection accuracy of minority class,a novel method was proposed.Based on the fact that the position of classification plane was determined by the boundary samples,the proposed method was implemented via removing some of samples closed to the classification plane to avoid the deficiency of the traditional SVM algorithm.Finally,the proposed method was compared with other approaches on unbalanced data sets.The simulation results show that the proposed method can not only increase the detection accuracy of minority but also improve the overall classification performance.

Key words: arrear, unbalance, support vector machine, boundary, under-sampling

CLC Number:

TP181

Chuangchuang LI,Guangyue LU,Hanglong WANG. SVM classifier for telecom user arrears based on boundary samples-based under-sampling approaches[J]. Telecommunications Science, 2017, 33(9): 85-91.

Figures/Tables 5

References 14

[1]	渠瑜 . 基于 SVM 的高不平衡分类技术研究及其在电信业的应用[D]. 杭州:浙江大学, 2010.
	QU Y . Research on SVM-based highly imbalanced classification and its application in telecommunications[D]. Hangzhou:Zhejiang University, 2010.
[2]	VAPNIK V N . The nature of statistical learning theory[M]. Berlin: SpringerPress, 2000: 138-167.
[3]	JAMIL S , KHAN A . Churn comprehension analysis for telecommunication industry using ALBA[C]// ICET,IEEE,October 18-19,2016,Islamabad,Pakistan. New Jersey:IEEE Press, 2016.
[4]	RASKUTTI B . Extreme re-balancing for SVMs:a case study[J]. ACM Sigkdd Explorations Newsletter, 2004,6(1): 60-69.
[5]	KANG P , CHO S . EUS SVMs:ensemble of under-sampled SVMs for data imbalance problems[M]. Berlin: SpringerPress, 2006: 837-846.
[6]	陶新民, 张冬雪, 郝思媛 ,等. 基于谱聚类下采样失衡数据下SVM故障检测[J]. 振动与冲击, 2013,32(16): 30-36.
	TAO X M , ZHANG D X , HAO S Y ,et al. Fault detection based on spectral clustering combined with under-sampling SVM under unbalanced datasets[J]. Journal of Vibration and Shock, 2013,32(16): 30-36.
[7]	HE H B , EDWARDO A . Learning from imbalanced data[J]. IEEE Transactions on Knowledge and Data Engineering, 2009,21(8): 1263-1284.
[8]	LIU X Y , ZHOU Z H . Exploratory under-sampling for class-imbalance learning[J]. IEEE Transactions on Systems,Man and Cybernetics, 2009,39(2): 539-550.
[9]	LIN C F , WANG S D . Fuzzy support vector machines[J]. IEEE Transactions on Neural Networks, 2002,13(2): 464-71.
[10]	HE H , GARCIA E A . Learning from imbalanced data[J]. IEEE Transactions on Knowledge ＆ Data Engineering, 2009,21(9): 1263-1284.
[11]	陶新民, 郝思媛, 张冬雪 ,等. 基于样本特性欠取样的不均衡支持向量机[J]. 控制与决策, 2013(7): 978-984.
	TAO X M , HAO S Y , ZHANG D X ,et al. Support vector machine for unbalanced data based on sample properties under-sampling approaches[J]. Control and Decision, 2013(7): 978-984.
[12]	BATUWITA R , PALADE V . FSVM-CIL:fuzzy support vector machines for class imbalance learning[J]. IEEE Transactions on Fuzzy Systems, 2010,18(3): 558-571.
[13]	李航 . 统计学习方法[M]. 北京: 清华大学出版社, 2012.
	LI H . Statistical learning method[M]. Beijing: Tsinghua University PressPress, 2012.
[14]	包志强, 崔妍 . 电信客户欠费模型评估[J]. 西安邮电大学学报, 2015(4): 97-101.
	BAO Z Q , CUI Y . Telecom customer arrearages model evaluation[J]. Journal of Xi’an University of Posts and Telecommunications, 2015(4): 97-101.

Metrics

Recommended 0

No Suggested Reading articles found!

样本属性	样本1	样本2	样本3	样本4	样本5
是否VIP	0	0	0	0	0
用户属性	0	0	1	0	1
付费方式	0	0	0	0	0
是否离网	0	0	0	0	0
是否主动离网	0	0	0	0	0
是否被动离网	0	0	0	0	0
是否主动停机	0	0	0	0	0
月通话时长/s	0	0	47 431	36 649	68 805
月通话次数/次	0	0	512	51	466
GPRS流量/KB	0	0	43 967	215 791	0
短信发送量/条	0	0	1	238	9
彩信发送量/条	1	2	20	20	23
套餐月费	1	5	2	2	2
通话费用/元	0	0	30.15	2.25	20.85
赠费销账/元	30	0	0	0	0
月初账费/元	42	193	131.85	157.27	150.75
账户余额/元	0	0	169.74	149.33	291.91

数据集	少数类数量	多数类数量	样本总数	不平衡比	类别数	正例类（第几类）
Haberman	81	225	306	26∶74	2	2
Pima-Indians	268	500	768	35∶65	2	1
Page-blocks	115	5 358	5 473	2∶98	5	5
Waveform	1 657	3 343	5 000	35∶65	2	1
abalone	103	4 074	4 177	2∶98	29	15
电信数据	250	3 000	3 250	8∶92	2	1

数据集	评价指标	B-SVM	标准SVM算法	模糊SVM算法	随机欠取样SVM算法	随机过取样SVM算法
Haberman	GM	0.68	0.420	0.547	0.592	0.624
	SE	0.55	0.198	0.325	0.625	0.475
	SP	0.84	0.893	0.82	0.56	0.82
Pima-Indians	GM	0.77	0.689	0.703	0.760	0.679
	SE	0.91	0.530	0.54	0.74	0.580
	SP	0.65	0.896	0.915	0.78	0.795
Page-blocks	GM	0.86	0.761	0.702	0.749	0.391
	SE	0.76	0.583	0.5	0.62	0.4
	SP	0.974	0.995	0.985	0.905	0.975
Waveform	GM	0.880	0.877	0.875	0.877	0.792
	SE	0.908	0.827	0.818	0.894	0.654
	SP	0.853	0.928	0.936	0.860	0.96
abalone	GM	0.49	0	0	0.539	0.02
	SE	0.46	0	0	0.58	0.02
	SP	0.523	1	1	0.5	0.997
电信数据	GM	0.931	0.561	0.918	0.871	0.871
	SE	0.92	0.32	0.88	0.88	0.8
	SP	0.942	0.983	0.958	0.862	0.948

[1]	Shitong LI, Daying QUAN, Zeyu TANG, Yun CHEN, Xiaofeng WANG, Xiaoping JIN. Time-frequency image and high-order spectrum characteristics based radar signal recognition [J]. Telecommunications Science, 2022, 38(2): 84-91.
[2]	Junning QIN,Jiajia HAN,Sheng ZHOU,Chunming WU,Shuangxi CHEN,Ruoyan ZHAO,Jiangyu ZHANG. Modeling of the mimic defense based on heterogeneous redundancy [J]. Telecommunications Science, 2020, 36(5): 31-38.
[3]	Yajie LI,Yongli ZHAO,Shoudong LIU,Jie ZHANG. Overview of research on fiber nonlinear equalization algorithm based on artificial intelligence [J]. Telecommunications Science, 2020, 36(3): 61-70.
[4]	Rongfang ZHANG,Dandan XU,Yuanguang WANG,Siyu PAN,Zhengmao LI. Application of machine learning in the fake user identification of IoT [J]. Telecommunications Science, 2019, 35(7): 136-144.
[5]	LEI Yuxiao,LI Gengfeng,HUANG Yuxiong,BIE Zhaohong. Support vector machine based fast Monte Carlo reliability evaluation method for composite power system [J]. Telecommunications Science, 2019, 35(6): 15-24.
[6]	Yaguan QIAN,Xiaohui GUAN,Shuhui WU,NBensheng YU,Dongxiao REN. A novel perturbation attack on SVM by greedy algorithm [J]. Telecommunications Science, 2019, 35(1): 81-89.
[7]	Xiao WANG,Rong ZHANG,Lijun GUO. Robust tampering detection and localization of composite image [J]. Telecommunications Science, 2017, 33(8): 120-129.
[8]	Zhenhong HE. Iris recognition based on generalized Gaussian distribution FDCT_Wrap and FSVM [J]. Telecommunications Science, 2016, 32(7): 126-131.
[9]	Yaguan QIAN,Xiaohui GUAN,Bensheng YUN,Qiong LOU,Pengfei MA. Internet traffic classification using SVM with flexible feature space [J]. Telecommunications Science, 2016, 32(5): 105-113.
[10]	Zhengjiang SONG,Xiaochen LI,Jiang CHEN. QoS boundary analysis of congestion control protocol based on stochastic network calculus in CRSN [J]. Telecommunications Science, 2016, 32(4): 44-51.
[11]	Zhenhong HE. Iris recognition algorithm based on feature weighting fast discrete Curvelet transform and fuzzy LS-SVM [J]. Telecommunications Science, 2016, 32(3): 92-98.
[12]	Hui Cheng,Jinglong Fang,Daquan Wang,Xingqi Wang. Performance Analysis of Simplification of Hyperplane Support Vector Machine [J]. Telecommunications Science, 2015, 31(8): 78-83.
[13]	Hao Dong,Hua Qu,Jihong Zhao,Liangjun Chen,Huijun Dai. Service-Aware Algorithm Based on DT-KSVM [J]. Telecommunications Science, 2015, 31(11): 67-71.
[14]	Ruitong Cao,Xu Cao. A Scheme of QoS Evaluation Based on Support Vector Machine [J]. Telecommunications Science, 2014, 30(2): 123-126.
[15]	Jiyuan Lu,Chenghui Huang,Fang Hou,Bin Li. Image Classification and Annotation Based on Hadoop Cloud Computing Platform [J]. Telecommunications Science, 2014, 30(2): 65-69.

SVM classifier for telecom user arrears based on boundary samples-based under-sampling approaches

RichHTML

PDF下载

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 5

References 14

Related Articles 15

Metrics

Recommended 0