一种针对不平衡数据分类的集成学习算法

doi:10.6040/j.issn.1672-3961.0.2019.050

摘要/Abstract

摘要：

针对水声目标识别中常被忽略的数据不平衡问题,提出一种随机子空间AdaBoost算法(RSBoost)。通过随机子空间法在不同水声特征空间上提取子训练样本集,在各个子训练样本集上训练基分类器,将其中少类间隔均值最大的基分类器作为本轮选定的分类器,迭代形成最终集成分类器。在实测数据上进行试验,利用F-measure和G-mean两个准则对RSBoost算法和AdaBoost算法在不同特征集上的性能进行评价。试验结果表明:相对于AdaBoost算法, RSBoost算法在F-measure准则下的平均值由0.07提升到0.22,在G-mean准则下的平均值由0.18提升到0.25,说明在处理水声数据不平衡分类问题上, RSBoost算法优于AdaBoost算法。

关键词: 不平衡数据, 集成学习, 水声目标识别, AdaBoost算法, 随机子空间

Abstract:

For unbalanced data classification problem in underwater acoustic target recognition, a random subspace AdaBoost algorithm called RSBoost was proposed. Subtraining sample set was extracted by random subspace method in different underwater acoustic feature space and base classifier was trained in every subtraining sample set. The base classifier with the maximum margin mean of minority class was chosen as the base classifier of this round, the final ensemble classifier was formed iteratively. The experiment was carried out on the measured data, the performance of RSBoost and AdaBoost in different feature space was evaluated by F-measure and G-mean. The results showed that, compared with AdaBoost, the F-measure of RSBoost improved from 0.07 to 0.22 and the G-mean improved from 0.18 to 0.25, which showed that RSBoost was superior to AdaBoost in underwater acoustic unbalanced classification problem.

Key words: unbalanced data, ensemble learning, underwater acoustic target recognition, AdaBoost algorithm, random space

中图分类号:

TP391

张宗堂,王森,孙世林. 一种针对不平衡数据分类的集成学习算法[J]. 山东大学学报 (工学版), 2019, 49(4): 8-13.

Zongtang ZHANG,Sen WANG,Shilin SUN. An ensemble learning algorithm for unbalanced data classification[J]. Journal of Shandong University(Engineering Science), 2019, 49(4): 8-13.

图/表 6

图1

表1

表2

图2

图3

表3

参考文献 13

1	DAI H L . Class imbalance learning via a fuuzy total margin based support vector machine[J]. Applied Soft Computing, 2015, 31, 172- 184. doi: 10.1016/j.asoc.2015.02.025
2	WANG S , YAO X . Using class imbalance learning for software defect prediction[J]. IEEE Trans on Reliability, 2013, 62 (2): 434- 443. doi: 10.1109/TR.2013.2259203
3	OZCIFT A , GULTEN A . Classifer ensemble construction with rotation forest to improve medical diagnosis performance of machine learning algorithms[J]. Computer Methods Programs Biomedicine, 2011, 104 (3): 443- 451. doi: 10.1016/j.cmpb.2011.03.018
4	KUBAT M, MATWIN S. Addressing the curse of imbalanced trainingsets: one-sided selection[C]//Proceedings of the Fourteenth International Conference on Machine Learning. New York, USA: Morgan Kaufmann, 1997: 179-186.
5	CHAWLA N , BOWYER K , HALL L , et al. SMOTE: synthetic minority over-sampling technique[J]. Journal of Artificial Intelligence Research, 2002, 16, 321- 357. doi: 10.1613/jair.953
6	张伶卫, 万文强. 基于云计算平台的代价敏感集成学习算法研究[J]. 山东大学学报(工学版), 2012, 42 (4): 19- 28.
	ZHANG Lingwei , WAN Wenqiang . Study on the cost-sensitive ensemble learning algorithm based on the cloud computing platform[J]. Journal of Shandong University(Engineering Science), 2012, 42 (4): 19- 28.
7	MANEVITZ L M , YOUSEFS M . One-class SVMs for document classification[J]. Journal of Machine Learning Research, 2001, 2, 139- 154.
8	李雄飞, 李军, 董元方, 等. 一种新的不平衡数据学习算法PCBoost[J]. 计算机学报, 2012, 35 (2): 2202- 2209.
	LI Xiongfei , LI Jun , DONG Yuanfang . A new learning algorithm for imbalanced data-PCBoost[J]. Chinese Journal of Computers, 2012, 35 (2): 2202- 2209.
9	BAIG M , AWAIS M , EL-ALFY Esm . AdaBoost-based artificial neural network learning[J]. Neurocomputing, 2017, 248 (1): 120- 126.
10	RCHAPIRE R , FREUND Y , BARLETT Y , et al. Boosting the margin: a new explanation for the effectiveness of voting methods[J]. The Annals of Statistics, 1998, 26 (5): 1651- 1686. doi: 10.1214/aos/1024691352
11	ZHOU Z H , WU J , TANG W . Ensembling neural networks: Many could be better than all[J]. Artificial Intelligence, 2002, 137 (1-2): 239- 263. doi: 10.1016/S0004-3702(02)00190-X
12	HO T K . The random subspace method for constructing decision forests[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1998, 20 (8): 832- 844. doi: 10.1109/34.709601
13	GAO Wei , ZHOU Zhihua . On the doubt about margin explanation of boosting[J]. Artificial Intelligence, 2013, 203, 1- 18. doi: 10.1016/j.artint.2013.07.002

相关文章 15

[1]	白琳,俱通,王浩,雷明珠,潘晓英. 面向不平衡数据的提升均衡集成学习算法[J]. 山东大学学报 (工学版), 2024, 54(4): 59-66.
[2]	常新功,苏敏惠,周志刚. 基于进化集成的图神经网络解释方法[J]. 山东大学学报 (工学版), 2024, 54(4): 1-12.
[3]	顾玉磊,马晖,王愚勤,胡卉,刘富鑫. 基于BP-Adaboost与TOPSIS的航空装备供应商评价[J]. 山东大学学报 (工学版), 2024, 54(1): 63-73.
[4]	闵海根,雷小平,李杰,童星,吴霞,方煜坤. 基于双层混合集成的自动驾驶汽车故障检测[J]. 山东大学学报 (工学版), 2022, 52(6): 30-40.
[5]	张大鹏,刘雅军,张伟,沈芬,杨建盛. 基于异质集成学习的虚假评论检测[J]. 山东大学学报 (工学版), 2020, 50(2): 1-9.
[6]	周荣翔,贾修一. 中文反语识别特征分析[J]. 山东大学学报 (工学版), 2019, 49(1): 41-46.
[7]	张璞,刘畅,王永. 基于特征融合和集成学习的建议语句分类模型[J]. 山东大学学报 (工学版), 2018, 48(5): 47-54.
[8]	沈冬冬,周风余,栗梦媛,王淑倩,郭仁和. 基于集成深度神经网络的室内无线定位[J]. 山东大学学报 (工学版), 2018, 48(5): 95-102.
[9]	王换,周忠眉. 一种基于聚类的过抽样算法[J]. 山东大学学报(工学版), 2018, 48(3): 134-139.
[10]	王立宏,李强. 旅行商问题的一种选择性集成求解方法[J]. 山东大学学报(工学版), 2016, 46(1): 42-48.
[11]	熊冰妍,王国胤,邓维斌. 分级式代价敏感决策树及其在手机换机预测中的应用[J]. 山东大学学报 (工学版), 2015, 45(5): 36-42.
[12]	陈大伟,闫昭*,刘昊岩. SVD系列算法在评分预测中的过拟合现象[J]. 山东大学学报(工学版), 2014, 44(3): 15-21.
[13]	鲁淑霞,李黎敏. 加权最大夹角间隔核心集向量机的不平衡数据分类[J]. 山东大学学报(工学版), 2014, 44(3): 1-7.
[14]	江伟坚1,2,郭躬德1,2*,赖智铭1,2. 基于新Haar-like特征的Adaboost人脸检测算法[J]. 山东大学学报(工学版), 2014, 44(2): 43-48.
[15]	李翔1,朱全银1,王尊2. 基于可变基函数和GentleAdaBoost的小波神经网络研究[J]. 山东大学学报(工学版), 2013, 43(5): 31-38.

多维度评价

Viewed

Full text

Abstract

Cited

Shared

Discussed

类别	预测少类	预测多类
实际少类	TP	FN
实际多类	FP	TN