受同轴对称抛物线约束的少数类样本合成方法

doi:10.19678/j.issn.1000-3428.0068963

摘要/Abstract

摘要： 线性插值是过采样技术中常见的样本合成方法,缺点是线性合成区域容易提高不同类别样本间的重叠度并降低采样结果的随机性,难以提升针对不平衡样本集的分类能力。为此,提出受同轴对称抛物线约束的样本合成方法。首先,建立一种自适应加权策略,通过危险度因子与相似度因子为少数类样本赋予权值,该权值能够决定采样过程中的样本合成方向及合成范围;然后,根据少数类样本与对应的样本权值构造一对同轴对称抛物线,并将这对抛物线相交得到的闭区域作为非线性的样本合成区域;最后,在引入新样本时通过该新样本近邻域内巴氏系数的变化情况判断此次采样是否能够有效避免入侵其他类别样本的分布区域,从而提高采样质量。在UCI机器学习库的6个公共样本集上实验结果表明,当C4.5作为分类器时,集成后的过采样方法与原始采样方法相比,精确率提高7.85百分点,召回率提高2.87百分点,G-means提高2.00百分点。

关键词: 线性插值, 抛物线, 自适应加权, 非线性合成区域, 巴氏系数

Abstract: Linear interpolation is often used by in over-sampling techniques to synthesize samples, but the its disadvantages include the lack of randomness in the sampling results and a tendency to increase the degree of class between samples of different categories, making it difficult to improve the classification ability for imbalanced sample sets. This paper proposes a generation method for minority samples with coaxial-symmetric parabolic constraints. First, for minority class samples, an adaptive weighting strategy based on the risk factor and similarity factor is established. The weight can determine the direction and range of sample synthesis during the sampling process. Then, a pair of coaxial symmetric parabolas based on minority class samples and corresponding sample weights is constructed, the closed region surrounded by a pair of coaxial symmetric parabolas is taken as the nonlinear synthesis region. Finally, when introducing a new sample, determine whether this sampling can effectively avoid invading the distribution areas of other categories of samples by observing the changes in the Bhattacharyya coefficient in the neighboring domain of the new sample, thereby improving the sampling quality. Comparison experiments on six public sample sets from the UCI show that when C4.5 is used as a classifier, the integrated oversampling method improves precision by 7.85 percentage points, recall by 2.87 percentage points, and G-means by 2.00 percentage points compared to the original sampling method.

Key words: linear interpolation, parabolas, adaptive weighting, nonlinear synthesis region, Bhattacharyya coefficient

中图分类号:

TP391

朱宸敏, 余粟. 受同轴对称抛物线约束的少数类样本合成方法[J]. 计算机工程, 2025, 51(5): 196-205.

ZHU Chenmin, YU Su. Synthesis Method for Minority Samples with Constraints of Coaxial Symmetric Parabolas[J]. Computer Engineering, 2025, 51(5): 196-205.

https://www.ecice06.com/CN/Y2025/V51/I5/196

参考文献

[1] HE H B,GARCIA E A. Learning from imbalanced data[J]. IEEE Transactions on Knowledge& Data Engineering, 2009, 21(9):1263-1284.
[2] 赵楠,张小芳,张利军.不平衡数据分类研究综述[J].计算机科学, 2018, 45(S1):22-27, 57. ZHAO N, ZHANG X F, ZHANG L J. Overview of imbalanced data classification[J]. Computer Science, 2018, 45(S1):22-27, 57.(in Chinese)
[3] YUAN X H, XIE L J, ABOUELENIEN M. A regularized ensemble framework of deep learning for cancer detection from multi-class,imbalanced training data[J]. Pattern Recognition, 2018, 77:160-172.
[4] YANG W J. Based on a semi-supervised fuzzy clustering and sample selection attribute reduction of the intrusion detection[C]//Proceedings of Conference on Mathematical, Computational and Statistical Sciences and Engineering. Washington D.C., USA:IEEE Press, 2016:193-197.
[5] CATENI S, COLLA V, VANNUCCI M. A method for resampling imbalanced datasets in binary classification tasks for real-world problems[J]. Neurocomputing, 2014, 135:32-47.
[6] WANG S, YAO X. Multiclass imbalance problems:analysis and potential solutions[J]. IEEE Transactions on Systems, Man, and Cybernetics. Part B (Cybernetics), 2012, 42(4):1119-1130.
[7] 刘定祥,乔少杰,张永清,等.不平衡分类的数据采样方法综述[J].重庆理工大学学报(自然科学), 2019, 33(7):102-112. LIU D X, QIAO S J, ZHANG Y Q, et al. A survey on data sampling methods in imbalance classification[J]. Journal of Chongqing University of Technology (Natural Science), 2019, 33(7):102-112.(in Chinese)
[8] 李昂,韩萌,穆栋梁,等.多类不平衡数据分类方法综述[J].计算机应用研究, 2022, 39(12):3534-3545. LI A, HAN M, MU D L, et al. Survey of multi-class imbalanced data classification methods[J].Application Research of Computers, 2022, 39(12):3534-3545.(in Chinese)
[9] CHAWLA N V, BOWYER K W, HALL L O, et al. SMOTE:synthetic minority over-sampling technique[J]. Journal of Artificial Intelligence Research, 2002, 16:321-357.
[10] HAN H, WANG W Y, MAO B H. Borderline-SMOTE:a new over-sampling method in imbalanced data sets learning[C]//Proceedings of the 2005 International Conference on Intelligent Computing. Berlin, Germany:Springer, 2005:878-887.
[11] HE H B,BAI Y, GARCIA E A, et al. ADASYN:adaptive synthetic sampling approach for imbalanced learning[C]//Proceedings of IEEE International Joint Conference on Neural Networks. Washington D. C., USA:IEEE Press, 2008:1322-1328.
[12] DOUZAS G, BACAO F, LAST F. Improving imbalanced learning through a heuristic oversampling method based on k-means and SMOTE[J].Information Sciences, 2018, 465:1-20.
[13] 姜新盈,王舒梵,严涛.基于层次密度聚类的去噪自适应混合采样[J].计算机系统应用, 2022, 31(10):206-210. JIANG X Y, WANG S F, YAN T. Denoising and adaptive hybrid sampling based on hierarchical density clustering[J]. Computer Systems and Applications, 2022, 31(10):206-210.(in Chinese)
[14] 韩明鸣,郭虎升,王文剑.面向非平衡多分类问题的二次合成QSMOTE方法[J].南京大学学报(自然科学), 2019, 55(1):1-13. HAN M M, GUO H S, WANG W J. Quadratic synthetic minority over-sampling technique for classification of multiclass imbalance problems[J]. Journal of Nanjing University (Natural Science), 2019, 55(1):1-13.(in Chinese)
[15] ZHU T F, LIN Y P, LIU Y H. Improving interpolation-based oversampling for imbalanced data learning[J]. Knowledge-Based Systems, 2020, 187:104826.
[16] 刘文英,林亚林,李克文,等.一种软件缺陷不平衡数据分类新方法[J].山东科技大学学报(自然科学版), 2021, 40(2):84-94. LIU W Y, LI Y L, LI K W, et al.A novel unbalanced data classification method for software defects[J]. Journal of Shandong University of Science and Technology (Natural Science), 2021,40(2):84-94.(in Chinese)
[17] 李国和,刘顺欣,张予杰,等.面向分类模型学习的样本类别均衡化方法[J].计算机应用与软件, 2022, 39(10):230-237. LI G H, LIU S X, ZHANG Y J, et al. Synthetic method of label-balancing samples for classifier learning[J]. Computer Applications and Software, 2022, 39(10):230-237.(in Chinese)
[18] 高子寒,宋燕.基于边界增强和去噪的自适应双权重过采样方法研究[J].智能计算机与应用, 2022, 12(1):58-64. GAO Z H, SONG Y. An adaptive double-weight enhanced boundary and denoising oversampling approach[J]. Intelligent Computer and Applications, 2022, 12(1):58-64.(in Chinese)
[19] 贺作伟,陶佳晴,冷强奎,等.带有超长方体约束的少数类样本生成机制[J].计算机应用研究, 2022, 39(10):3055-3060. HE Z W, TAO J Q, LENG Q K, et al. Generation mechanism for minority samples with hypercuboid constraints[J]. Application Research of Computers, 2022, 39(10):3055-3060.(in Chinese)
[20] 贺永森,陈江.抛物线插值法用于液压机工艺曲线的研究[J].锻压装备与制造技术, 2022, 57(4):68-71. HE Y S,CHEN J. Study on process curve of hydraulic press by use of parabolic interpolation[J].China Metalforming Equipment& Manufacturing Technology, 2022, 57(4):68-71.(in Chinese)
[21] 马兰,井伟,扈月松,等.两点抛物线插值提高雷达测距精度的研究[J].火控雷达技术, 2020, 49(4):14-18, 26. MA L, JING W, HU Y S, et al. Research on improving radar ranging accuracy by two-point parabolic interpolation[J]. Fire Control Radar Technology, 2020, 49(4):14-18,26.(in Chinese)
[22] 董明刚,姜振龙,敬超.基于海林格距离和SMOTE的多类不平衡学习算法[J].计算机科学, 2020, 47(1):102-109. DONG M G, JIANG Z L, JING C. Multi-class imbalanced learning algorithm based on Hellinger distance and SMOTE algorithm[J]. Computer Science, 2020, 47(1):102-109.(in Chinese)
[23] 吴煜,杨爱萍,章宦记,等.基于黎曼与巴氏距离的脑磁图信号分类方法[J].计算机科学与探索, 2017, 11(5):776-784. WU Y, YANG A P, ZHANG H J, et al. MEG signals classification algorithm based on Riemann and Bhattacharyya distances[J]. Journal of Frontiers of Computer Science and Technology, 2017, 11(5):776-784.(in Chinese)
[24] 李敏波,董伟伟.面向不平衡数据集的汽车零部件质量预测[J].中国机械工程, 2022, 33(1):88-96. LI M B, DONG W W. Quality prediction of automotive parts for imbalanced datasets[J]. China Mechanical Engineering, 2022, 33(1):88-96.(in Chinese)
[25] LICHMAN M. UCI machine learning repository[EB/OL].[2023-11-02] . http://archive.ics.uci.edu/ml.

选择文件类型/文献管理软件名称

选择包含的内容