具有标签流形和动态图约束的多标签特征选择

doi:10.6040/j.issn.1671-9352.7.2024.452

摘要/Abstract

摘要： 将自适应动态图技术和标签流形集成到改进后的线性映射学习框架中,提出了具有标签流形和动态图约束的多标签特征选择算法。该算法基于特征自表示的改进矩阵分解技术,改进了线性映射模型,对特征和标签之间以及不同标签之间的相关性进行解耦。设计了一种具有拉普拉斯秩约束的自适应动态图技术,学习高质量的特征相似图。构建了基于标签相关性的标签流形,将标签信息充分的纳入模型的训练中。验证了自适应动态图技术可以有效的提高图矩阵的质量,以及所提算法在解决多标签特征选择问题上的有效性。

关键词: 多标签学习, 特征选择, 流形学习, 自适应学习, 动态图学习

Abstract: Multi-label feature selection algorithms with label manifolds and dynamic graph constraints are proposed by integrating adaptive dynamic graph technique and label manifold into an improved linear mapping learning framework. In this algorithm, an improved matrix decomposition technique based on feature self-representation improves the linear mapping model and decouples the correlation between features and labels as well as between different labels. An adaptive dynamic graph technique with Laplace rank constraints is designed to learn a high-quality feature similarity graph. A label manifold based on label relevance is constructed to fully incorporate label information into the training of the algorithm. Numerous experimental results verify that the adaptive dynamic graph technique can effectively improve the quality of the graph matrix and the effectiveness of the proposed algorithm in addressing the multi-label feature selection problem.

Key words: multi-label learning, feature selection, manifold learning, adaptive learning, dynamic graph learning

中图分类号:

TP181

武晓军,陈怡丹,郝耀军,宋长伟,何德清. 具有标签流形和动态图约束的多标签特征选择[J]. 《山东大学学报(理学版)》, 2025, 60(7): 69-83.

WU Xiaojun, CHEN Yidan, HAO Yaojun, SONG Changwei, HE Deqing. Multi-label feature selection with label manifold and dynamic graph constraints[J]. JOURNAL OF SHANDONG UNIVERSITY(NATURAL SCIENCE), 2025, 60(7): 69-83.

参考文献

[1] WANG Boyan, HU Xuegang, LI Peipei, et al. Cognitive structure learning model for hierarchical multi-label text classification[J]. Knowledge-Based Systems, 2021, 218:106876.
[2] XIONG Jie, YU Li, NIU Xi, et al. XRR: extreme multi-label text classification with candidate retrieving and deep ranking[J]. Information Sciences, 2023, 622:115-132.
[3] KOMEILI M, LOUIS W, ARMANFARD N, et al. Feature selection for nonstationary data:application to human recognition using medical biometrics[J]. IEEE Transactions on Cybernetics, 2017, 48(5):1446-1459.
[4] JANET J P, KULIK H J. Resolving transition metal chemical space:feature selection for machine learning and structure-property relationships[J]. The Journal of Physical Chemistry A, 2017, 121(46):8939-8954.
[5] BERMINGHAM M L, PONG-WONG R, SPILIOPOULOU A, et al. Application of high-dimensional feature selection: evaluation for genomic prediction in man[J]. Scientific Reports, 2015, 5:10312.
[6] HASTIE T, TIBSHIRANI R, FRIEDMAN J. The elements of statistical learning:data mining, inference, and prediction[J]. The Mathematical Intelligencer, 2005, 27(2):83-85.
[7] SUN Xin, LIU Yanheng, LI Jin, et al. Using cooperative game theory to optimize the feature selection problem[J]. Neurocomputing, 2012, 97:86-93.
[8] KASHEF S, NEZAMABADI-POUR H, NIKPOUR B. Multilabel feature selection:a comprehensive review and guiding experiments[J]. Wiley Interdisciplinary Reviews: Data Mining and Knowledge Discovery, 2018, 8(2):e1240.
[9] YIN Hui, YANG Shuiqiao, SONG Xiangyu, et al. Deep fusion of multimodal features for social media retweet time prediction[J]. World Wide Web, 2021, 24:1027-1044.
[10] SHI Jianbo, MALIK J. Normalized cuts and image segmentation[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2000, 22(8):888-905.
[11] ROWEIS S T, SAUL L K. Nonlinear dimensionality reduction by locally linear embedding[J]. Science, 2000, 290(5500):2323-2326.
[12] ZHANG Yao, MA Yingcang, YANG Xiaofei. Multi-label feature selection based on logistic regression and manifold learning[J]. Applied Intelligence,2022, 52(8):9256-9273.
[13] ZHANG Yao, MA Yingcang. Sparse multi-label feature selection via dynamic graph manifold regularization[J]. International Journal of Machine Learning and Cybernetics, 2023, 14(3):1021-1036.
[14] HU Juncheng, LI Yonghao, XU Gaochao, et al. Dynamic subspace dual-graph regularized multi-label feature selection[J]. Neurocomputing, 2022, 467:184-196.
[15] 李永豪,胡亮,张平,等. 基于动态图拉普拉斯的多标签特征选择[J]. 通信学报,2020,41(12):47-59. LI Yonghao, HU Liang, ZHANG Ping, et al. Multi-label feature selection based on dynamic graph Laplacian[J]. Journal on Communications, 2020, 41(12):47-59.
[16] ZHANG Yao, MA Yingcang. Non-negative multi-label feature selection with dynamic graph constraints[J]. Knowledge-Based Systems, 2022, 238:107924.
[17] LI Yonghao, HU Liang, GAO Wanfu. Robust sparse and low-redundancy multi-label feature selection with dynamic local and global structure preservation[J]. Pattern Recognition, 2023, 134:109120.
[18] ZHANG Minling, ZHOU Zhihua. A review on multi-label learning algorithms[J]. IEEE Transactions on Knowledge and Data Engineering, 2013, 26(8):1819-1837.
[19] BOUTELL M R, LUO Jiebo, SHEN Xipeng, et al. Learning multi-label scene classification[J]. PatternRecognition, 2004, 37(9):1757-1771.
[20] HUANG Jun, LI Guorong, HUANG Qingming, et al. Learning label specific features for multi-label classification[C] // Proceedings of the 2015 IEEE International Conference on Data Mining. Atlantic City, New Jersey, IEEE, 2015:181-190.
[21] ZHANG Jia, LUO Zhiming, LI Candong, et al. Manifold regularized discriminative feature selection for multi-label learning[J]. Pattern Recognition, 2019, 95:136-150.
[22] HU Juncheng, LI Yonghao, GAO Wanfu, et al. Robust multi-label feature selection with dual-graph regularization[J]. Knowledge-Based Systems, 2020, 203:106126.
[23] LI Yonghao, HU Liang, GAO Wanfu. Multi-label feature selection via robust flexible sparse regularization[J]. Pattern Recognition, 2023, 134:109074.
[24] CAI Zhiling, ZHU William. Multi-label feature selection via feature manifold learning and sparsity regularization[J]. International Journal of Machine Learning and Cybernetics, 2018, 9:1321-1334.
[25] GAO Wanfu, LI Yonghao, HU Liang. Multilabel feature selection with constrained latent structure shared term[J]. IEEE Transactions on Neural Networks and Learning Systems, 2023, 34(3):1253-1262.
[26] LI Yonghao, HU Liang, GAO Wanfu. Label correlations variation for robust multi-label feature selection[J]. Information Sciences, 2022, 609:1075-1097.
[27] HAN Jiuqi, SUN Zhengya, HAO Hongwei. Selecting feature subset with sparsity and low redundancy for unsupervised learning[J]. Knowledge-Based Systems, 2015, 86:210-223.
[28] ZHANG Yao, HUO Wei, TANG Jun. Multi-label feature selection via latent representation learning and dynamic graph constraints[J]. Pattern Recognition, 2024, 151:110411.
[29] LIN Yaojin, HU Qinghua, LIU Jinghua, et al. Multi-label feature selection based on max-dependency and min-redundancy[J]. Neurocomputing, 2015, 168:92-103.
[30] LEE J, KIM D W. SCLS: multi-label feature selection based on scalable criterion for large label set[J]. Pattern Recognition, 2017, 66:342-352.
[31] HASHEMI A, DOWLATSHAHI M B, NEZAMABADI-POUR H. MFS-MCDM: multi-label feature selection using multi-criteria decision making[J]. Knowledge-Based Systems, 2020, 206:106365.
[32] ZOU Yizhang, HU Xuegang, LI Peipei. Gradient-based multi-label feature selection considering three-way variable interaction[J]. Pattern Recognition, 2024, 145:109900.
[33] XU Wei, GONG Yihong. Document clustering by concept factorization[C] // Proceedings of the 27thAnnual International ACM SIGIR Conference on Research and Development in Information Retrieval. Sheffield: ACM, 2004:202-209.
[34] FAN K. On a theorem of Weyl concerning eigenvalues of linear transformations I [J]. Proceedings of the National Academy of Sciences, 1949, 35(11):652-655.
[35] LEE D D, SEUNG H S. Learning the parts of objects by non-negative matrix factorization[J]. Nature, 1999, 401(6755):788-791.
[36] HUANG Jin, NIE Feiping, HUANG Heng. A new simplex sparse learning model to measure data similarity for clustering[C] // Proceedings of the 24th International Joint Conference on Artificial Intelligence. Burlington: Morgan Kaufmann, 2015:3569-3575.
[37] LEE D D, SEUNG S H. Algorithms for non-negative matrix factorization[J]. Advances in Neural Information Processing Systems, 2001, 13:556-562.
[38] DING C H Q, Li Tao, JORDAN M I. Convex and semi-nonnegative matrix factorizations[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2008, 32(1):45-55.
[39] ZHANG Minling, ZHOU Zhihua. ML-KNN: a lazy learning approach to multi-label learning[J]. Pattern Recognition, 2007, 40(7):2038-2048.
[40] DOUGHERTY J, KOHAVI R, SAHAMI M. Supervised and unsupervised discretization of continuous features[M].Massachusetts: Morgan Kaufmann, 1995.
[41] XUE Guotong, ZHONG Ming, LI Jianxin, et al. Dynamic network embedding survey[J]. Neurocomputing, 2022, 472:212-223.
[42] DUNN O J. Multiple comparisons among means[J]. Journal of the American statistical association, 1961, 56(293):52-64.
[43] FRIEDMAN M. A comparison of alternative tests of significance for the problem ofm rankings[J]. The Annals of Mathematical Statistics, 1940, 11(1):86-92.

多维度评价

Viewed

Full text

Abstract

Cited

Shared

Discussed