用于处理不平衡样本的改进近似支持向量机新算法

doi:10.11772/j.issn.1001-9081.2014.06.1618

计算机应用 ›› 2014, Vol. 34 ›› Issue (6): 1618-1621.DOI: 10.11772/j.issn.1001-9081.2014.06.1618

用于处理不平衡样本的改进近似支持向量机新算法

刘艳¹,²,钟萍²,陈静²,宋晓华¹,何云³

1. 燕京理工学院机电学院,河北廊坊 065201
2. 中国农业大学理学院,北京 100083
3.

收稿日期:2013-11-18 修回日期:2014-01-21 发布日期:2014-07-02 出版日期:2014-06-01
通讯作者: 陈静
作者简介:刘艳(1984-),女,湖南娄底人,助教,硕士,主要研究方向:数据挖掘、支持向量机;钟萍(1971-),女,山东青岛人,教授,博士,主要研究方向:支持向量机;陈静(1964-),女,河南南阳人,教授, 主要研究方向:数据挖掘、支持向量机、最优化方法、数值模拟。
基金资助:
国家自然科学基金资助项目

Modified proximal support vector machine algorithm for dealing with unbalanced samples

LIU Yan¹,²,ZHONG Ping²,CHEN Jing²,SONG Xiaohua¹,HE Yun¹

1. College of Mechanical and Electrical Engineering, Yanching Institute of Technology, Langfang Hebei 065201, China
2. College of Science, China Agricultural University, Beijing 100083, China;

Received:2013-11-18 Revised:2014-01-21 Online:2014-07-02 Published:2014-06-01
Contact: CHEN Jing

摘要/Abstract

摘要：

近似支持向量机(PSVM)在处理不平衡样本时,会过拟合样本点数较多的一类,低估样本点数较少的类的错分误差,从而导致整体样本的分类准确率下降。针对该问题，提出一种用于处理不平衡样本的改进的PSVM新算法。新算法不仅给正、负类样本赋予不同的惩罚因子,而且在约束条件中新增参数,使得分类面更具灵活性。该算法先对训练集训练获得最优参数,然后再对测试集进行训练获得分类超平面,最后输出分类结果。UCI数据库中9组数据集的实验结果表明:新算法提高了样本的分类准确率,在线性的情况下平均提高了2.19个百分点,在非线性的情况下平均提高了3.14个百分点,有效地提高了模型的泛化能力。

Abstract:

When Proximal Support Vector Machine (PSVM) deals with unbalanced samples, it will overfit the class with large samples and underestimate the misclassification error of the class with small samples, resulting in the decline of accuracy in overall samples. To solve this problem, a modified PSVM used for dealing with unbalanced samples was proposed. The new algorithm not only set different punishments for positive and negative samples, but also added a new parameter to the constraint, making the classification hyperplane more flexible. Firstly, the new algorithm trained the training set to obtain the optimal parameters, then the classification hyperplane was obtained by training the test set. Finally, the classification results was output. The experiments presented by 9 datasets in UCI database show that the new algorithm improves the classification accuracy of the samples, by 2.19 and 3.14 percentage points in the linear and nonlinear case respectively. The generalization ability of the algorithm is strengthened effectively.

中图分类号:

TP18

刘艳钟萍陈静宋晓华何云. 用于处理不平衡样本的改进近似支持向量机新算法[J]. 计算机应用, 2014, 34(6): 1618-1621.

LIU Yan ZHONG Ping CHEN Jing SONG Xiaohua HE Yun. Modified proximal support vector machine algorithm for dealing with unbalanced samples[J]. Journal of Computer Applications, 2014, 34(6): 1618-1621.

参考文献

[1]BRADLEY P S, MANGASARIAN O L. Massive data discrimination via linear support vector machines [J]. Optimization Methods and Software, 2000,13(1): 1-10.
[2]BURGES C J C. A tutorial on support vector machines for pattern recognition [J]. Data Mining and Knowledge Discovery, 1998, 2(2): 121-167.
[3]VAPNIK V N. The nature of statistical learning theory[M]. Berlin: Springer-Verlag, 1995.
[4]FUNG G, MANGASARIAN O L. Proximal support vector machine classifiers [C]// KDD 2001: Proceedings of the 7th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. New York: ACM Press, 2001 :77-82.
[5]FUNG G, MANGASARIAN O L. Incremental support vector machine classification[C]// Proceedings of the 2nd SIAM International Conference on Data Mining. Philadelphia: University of Wisconsin,2002 : 247-260.
[6]AGARWAL D K. Shrinkage estimator generalizations of proximal support vector machine[C]// Proceedings of the 8th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. New York: ACM Press, 2002: 173-182.
[7]LI K, HUANG H. Incremental learning proximal support vector machine classifier [C]// Proceedings of the 2002 International Conference on Machine Learning and Cybernetics. Piscataway: IEEE Press, 2002,3:1635-1637.
[8]YU X, LU W, CHU F. Rotating machinery fault diagnosis based on fuzzy proximal support vector machine optimized by particle swarm optimization[J]. Journal of Vibration and Shock, 2009,28(11): 183-186.(于湘涛,卢文秀,褚福磊. 基于PSO算法的模糊PSVM及其在旋转机械故障诊断中的应用[J]. 振动与冲击,2009,28(11): 183-186.)
[9]WANG Y, GAO X, LI J. A PSVM-based active learning method for mass detection[J]. Journal of Computer Research and Development, 2012,49(3):572-578.(王颖,高新波,李洁. 基于PSVM的主动学习肿块检测方法[J]. 计算机研究与发展,2012,49(3):572-578.)
[10]FUNG G, MANGASARIAN O L. Multicategory proximal support vector machine classifiers [J ]. Machine Learning, 2005,59(1/2):77-97.
[11]TAO X, JI H, DONG S. Modified PSVM and its application in unbalanced data classification [J]. Computer Engineering, 2007, 33(24):191-193.(陶晓燕,姬红兵,董淑福. 改进的PSVM及其在非平衡数据分类中的应用[J]. 计算机工程,2007, 33(24):191-193.)
[12]ZHUANG D, CHEN Y. Text classification by weighted proximal support vector machine [J]. Journal of Tsinghua University: Science and Technology, 2005,45(S1):1787-1790.(庄东, 陈英. 基于加权近似支持向量机的文本分类[J]. 清华大学学报:自然科学版,2005,45(S1): 1787-1790.)
[13]ZHANG M, FU L, WANG G. Fuzzy proximal support vector machine [J]. Computer Engineering and Applications, 2005,41(5):37-39.(张猛, 付丽华, 王高峰. 模糊临近支持向量机[J]. 计算机工程与应用, 2005,41(5):37-39.)
[14]WANG X, CUI F, LU S. Density weighted proximal support vector machine[J].Computer Science, 2012,39(1):182-184. (王熙照,崔芳芳,鲁淑霞. 密度加权近似支持向量机[J]. 计算机科学,2012,39(1):182-184.)
[15]WU Z, DOU H. Generalized least squares support vector machine algorithm and its application [J]. ,2009, 29(3):877-879.(吴宗亮,窦衡.一种广义最小二乘支持向量机算法及其应用[J]. 计算机应用,2009,29(3):877-879.)

用于处理不平衡样本的改进近似支持向量机新算法

Modified proximal support vector machine algorithm for dealing with unbalanced samples

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

[1]	吕乐张博瀚荆军昌刘栋. 基于持久性的多目标节点隐藏方法[J]. 《计算机应用》唯一官方网站, 0, (): 0-0.
[2]	宫智宇王士同. 面向重尾噪声图像分类的残差网络学习方法[J]. 《计算机应用》唯一官方网站, 0, (): 0-0.
[3]	王虎王晓峰李可马云洁. 融合多头自注意力的标签语义嵌入联邦类增量学习方法[J]. 《计算机应用》唯一官方网站, 0, (): 0-0.
[4]	丁建立, 黄辉, 曹卫东. 航班链运行状态动态监控方法[J]. 《计算机应用》唯一官方网站, 2024, 44(12): 3941-3948.
[5]	刘晶鑫, 黄雯静, 徐亮胜, 黄冲, 吴建生. 字典学习与样本关联保持结合的无监督特征选择模型[J]. 《计算机应用》唯一官方网站, 2024, 44(12): 3766-3775.
[6]	沈嫣然, 温昕, 张瑾昊, 张帅, 曹锐, 高保禄. 轻量级多尺度卷积网络的功能磁共振成像脑龄预测模型[J]. 《计算机应用》唯一官方网站, 2024, 44(12): 3949-3957.
[7]	张祖篡, 陈学斌, 高瑞, 邹元怀. 基于标签分类的联邦学习客户端选择方法[J]. 《计算机应用》唯一官方网站, 2024, 44(12): 3759-3765.
[8]	宋逸飞, 柳毅. 基于数据增强和标签噪声的快速对抗训练方法[J]. 《计算机应用》唯一官方网站, 2024, 44(12): 3798-3807.
[9]	蒋权黄文清苟志勇. 基于等变图神经网络的拉格朗日粒子流模拟[J]. 《计算机应用》唯一官方网站, 0, (): 0-0.
[10]	李岚皓严皓钧周号益孙庆赟李建欣. 基于神经网络的多尺度信息融合时间序列长期预测模型[J]. 《计算机应用》唯一官方网站, 0, (): 0-0.
[11]	廖炎华鄢元霞潘文林. 基于YOLOv9的交通路口图像的多目标检测算法[J]. 《计算机应用》唯一官方网站, 0, (): 0-0.
[12]	张学飞张丽萍闫盛侯敏赵宇博. 知识图谱与大语言模型协同的个性化学习推荐[J]. 《计算机应用》唯一官方网站, 0, (): 0-0.
[13]	索晋贤张丽萍闫盛王东奇张雅雯. 可解释的深度知识追踪方法综述[J]. 《计算机应用》唯一官方网站, 0, (): 0-0.
[14]	陈丹阳张长伦. 多尺度去相关的图卷积网络[J]. 《计算机应用》唯一官方网站, 0, (): 0-0.
[15]	蒋沛宇王永光任亚亭李硕晨谭火彬. 基于测量不确定度表示指南的目标检测不确定度测量方案[J]. 《计算机应用》唯一官方网站, 0, (): 0-0.