[1]秦晓燕,郑晓晗,张莉.基于局部邻接图的半监督稀疏回归算法[J].南京师大学报(自然科学版),2025,48(04):96-105.[doi:10.3969/j.issn.1001-4616.2025.04.010]
　Qin Xiaoyan,Zheng Xiaohan,Zhang Li.Semi-Supervised Sparse Regression Based on Local Adjacency Graph[J].Journal of Nanjing Normal University(Natural Science Edition),2025,48(04):96-105.[doi:10.3969/j.issn.1001-4616.2025.04.010]
点击复制

基于局部邻接图的半监督稀疏回归算法()

《南京师大学报（自然科学版）》[ISSN:1001-4616/CN:32-1239/N]

卷:
48
期数:
2025年04期
页码:
96-105
栏目:
计算机科学与技术
出版日期:
2025-08-20

文章信息/Info

Title:
Semi-Supervised Sparse Regression Based on Local Adjacency Graph
文章编号:
1001-4616(2025)04-0096-10
作者:
秦晓燕1; 郑晓晗2; 张莉1; 2
(1.苏州高博职业学院信息与软件学院,江苏 苏州 215163)
(2.苏州大学计算机科学与技术学院,江苏 苏州 215006)
Author(s):
Qin Xiaoyan1; Zheng Xiaohan2; Zhang Li1; 2
(1.School of Information and Software,Suzhou Global Institute,Suzhou 215163,China)
(2.School of Computer Science and Technology,Soochow University,Suzhou 215006,China)
关键词:
半监督回归学习; 图生成; 邻接矩阵; 稀疏性; 1范数正则
Keywords:
semi-supervised regression learning; adjacency matrix; graph generation; sparsity; 1-norm regularization
分类号:
TP181
DOI:
10.3969/j.issn.1001-4616.2025.04.010
文献标志码:
A
摘要:
半监督回归算法可以利用少量有标签样本和大量无标签样本进行回归建模,进而在一定程度上解决了获取标签样本成本高的问题. 目前,已经提出了能利用邻接矩阵来挖掘数据潜在结构的图半监督回归算法. 然而,这些方法存在两个问题. 第一,现有半监督回归算法多使用全连接的图生成方式,很容易受到噪声或离群点的影响; 第二,现有图半监督回归算法的稀疏性不足. 为了解决上述问题,本文提出了一种新的半监督回归学习算法——基于局部邻接图的半监督稀疏回归算法. 该算法构建了一种新的局部邻接图,在生成邻接矩阵时仅关注样本的局部信息,从而保留了数据的局部流形结构,缓解噪声样本对算法的影响. 另外,利用1范数正则能诱导稀疏性的特性,本文在优化问题中引入模型系数的1范数正则项,有效地提高了模型的稀疏性. 本文在9个真实数据集上对算法的半监督回归性能和稀疏性进行了实验验证. 实验结果表明,本文所提算法在不同实验设置下均能获得较好的回归性能.
Abstract:
Semi-supervised regression(SSR)algorithms leverage a small amount of labeled samples along with a large pool of unlabeled samples for modeling regression functions,which can alleviate the high costs associated with obtaining labeled data to some extent. At present,graph-based SSR algorithms have been proposed,which can employ adjacency matrices to dig the underlying data structure. However,existing graph-based SSR methods are confronted with two primary challenges. First,existing methods rely on the fully-connected graph on data that is susceptible to noise and outliers. Second,the sparsity of current SSR algorithms leaves room for enhancement. To address these issues,this paper proposes a novel semi-supervised sparse regression algorithm. This algorithm designs a new graph,the local adjacency graph,which focuses on the local connectivity of samples. This graph generation method preserves the local manifold structure of the data and mitigates the impact of noisy samples. Furthermore,our algorithm capitalizes on the sparsity-inducing property of the 1-norm regularization by incorporating a 1-norm regularization term for the model coefficients into the optimization problem,effectively enhancing the model's sparsity. Empirical validation is performed on nine real-world datasets to evaluate the SSR performance and the sparsity of the proposed algorithm. Results demonstrate that our method achieves superior performance across various experimental setups.

参考文献/References:

[1]VAN ENGELEN J E,HOOS H H. A survey on semi-supervised learning[J]. Machine learning,2020,109(2):373-440.
[2]邓廷权,王强. 半监督类保持局部线性嵌入方法[J]. 智能系统学报,2021,16(1):98-107.
[3]翟德明,沈斯娴,周雄,等. 基于最优传输理论的深度半监督学习伪标签生成算法[J]. 软件学报:1-14[2024-06-28]. https://doi.org/10.13328/j.cnki.jos.007054.
[4]胡声丹,苗夺谦,姚一豫. 基于三支标签传播的半监督属性约简[J]. 计算机学报,2021,44(11):2332-2343.
[5]KOSTOPOULOS G,KARLOS S,KOTSIANTIS S,et al. Semi-supervised regression:a recent review[J]. Journal of intelligent & fuzzy systems,2018,35(2):1483-1500.
[6]郭兰哲,李宇峰. 稳健选择伪标注的混合式半监督学习[J]. 中国科学:信息科学,2024,54(3):623-637.
[7]AZRIEL D,BROWN L D,SKLAR M,et al. Semi-supervised linear regression[J]. Journal of the American Statistical Association,2022,117(540):2238-2251.
[8]CHEN X,CAO W,GAN C,et al. Semi-supervised support vector regression based on data similarity and its application to rock-mechanics parameters estimation[J]. Engineering applications of artificial intelligence,2021,104:104317.
[9]杨智勇,许倩倩,何源,等. 半监督AUC优化的Boosting算法及理论[J]. 计算机学报,2022,45(8):1598-1617.
[10]ZHOU Z,LI M. Semi-supervised regression with cotraining-style algorithms[J]. IEEE transactions on knowledge and data engineering,2007,19(11):1479-1493.
[11]ZHOU Z,LI M. Semi-supervised regression with co-training[C]//Proceedings of International Joint Conference on Artificial Intelligence. San Francisco:Margan Kaufmann,2005:908-913.
[12]YOO J,KIM H J. Semisupervised location awareness in wireless sensor networks using Laplacian support vector regression[J]. International journal of distributed sensor networks,2014,10(4):265801.
[13]YANG H,QIN F,LIANG Q L,et al. LapRLSR for NIR spectral modeling and its application to online monitoring of the column separation of Salvianolate[J]. Chinese chemical letters,2007,18(7):852-856.
[14]ZHANG L,ZHOU W. On the sparseness of 1-norm support vector machines[J]. Neural networks,2010,23(3):373-385.
[15]郑晓晗. 稀疏与鲁棒的孪生支持向量机算法研究[D]. 苏州:苏州大学,2021.
[16]ZHENG X,ZHANG L,XU Z. L1-norm Laplacian support vector machine for data reduction in semi-supervised learning[J]. Neural computing and applications,2023,35(17):12343-12360.
[17]ZHENG X,ZHANG L,YAN L. Sparse discriminant twin support vector machine for binary classification[J]. Neural computing and applications,2022,34(19):16173-16198.
[18]WANG J,GUO S,NIE F,LI X. Local-global fuzzy clustering with anchor graph[J]. IEEE transactions on fuzzy systems,2024,32(1):188-202.
[19]ZHU J,CHEN X,YANG H,et al. Unsupervised adaptive bipartite graph embedding[J]. IEEE transactions on knowledge and data engineering,2023,35(10):10514-10525.
[20]CHEN L,ZHONG Z. Adaptive and structured graph learning for semi-supervised clustering[J]. Information processing & management,2022,59(4):102949.
[21]李静,陈秀宏. 基于竞争性协同表示的局部判别投影特征提取[J]. 智能系统学报,2019,14(5):974-981.
[22]SMOLA A J,SCHOLKOPF B. A tutorial on support vector regression[J]. Statistics and computing,2004,14:199-222.
[23]ROWEIS S T,SAUL L K. Nonlinear dimensionality reduction by locally linear embedding[J]. Science,2000,290(5500):2323-2326.
[24]陈聪. L1正则化与pinball损失函数的极限学习机[J]. 信息技术与信息化,2023(3):37-40.
[25]KWAK N. Principal component analysis based on L1-norm maximization[J]. IEEE transactions on pattern analysis and machine intelligence,2008,30(9):1672-1680.
[26]WANG H,LU X,HU Z,ZHENG W. Fisher discriminant analysis with L1-norm[J]. IEEE transactions on cybernetics,2013,44(6):828-842.
[27]杨红鑫,杨绪兵,寇振宇,等. 基于L1范数的k平面聚类算法设计[J]. 南京航空航天大学学报,2019,51(5):681-686.
[28]WANG X,MA L,WANG X. Apply semi-supervised support vector regression for remote sensing water quality retrieving[C]//Proceedings of IEEE International Geoscience and Remote Sensing Symposium,Piscataway,NJ:IEEE,2010:2757-2760.
[29]HUA Z,YANG Y,QIU H. Node influence-based label propagation algorithm for semi-supervised learning[J]. Neural computing and applications,2020,33(7):2753-2768.
[30]ZHANG L,ZHOU W,CHANG P,et al. Kernel sparse representation-based classifier[J]. IEEE transactions on signal processing,2011,60(4):1684-1695.

备注/Memo

备注/Memo:
收稿日期:2024-09-24.
基金项目:江苏省高校自然科学研究资助项目(19KJA550002)、江苏省六大人才高峰资助项目(XYDXX-054)、江苏省职业教育软件技术“双师型”名师工作室资助项目(苏教师函[2022]31号).
通讯作者:张莉,教授,博士生导师,研究方向:机器学习. E-mail:zhangliml@suda.edu.cn
更新日期/Last Update: 2025-08-20