Abstract
Multi-label classification encounters great challenges in processing high dimensional input feature and output label spaces. In this paper, we propose a geometrically preserved dual projections learning of both label and feature spaces for multi-label classification. Different from traditional multi-label classification algorithms which learn appropriate feature dimensions directly from feature space, our proposed method learns dual label and feature space projections from both input feature and output label spaces simultaneously. Thus, achieving dimension reduction in both label and feature spaces. Furthermore, since labels and features are sampled from label and feature manifolds, respectively, we learn their geometric structures by constructing label and feature graphs. By geometrically preserving dual projections, better low rank structures in label and feature spaces are obtained, which further improves the performance of multi-label classification. Experimental results on several data sets such as cal500, enron, languagelog and rcv1s1, demonstrate the proposed method outperforms the state-of-the-art multi-label classification methods, such as MIFS, RAKEL, PLST, CPLST, GroPLE, CSSP, FaIE in LSDR and STOA.
Similar content being viewed by others
References
Schapire RE, Singer Y (2004) Boostexter: a boosting-based system for text categorization. Mach Learn 39:135–168
Liu Y, Wen K, Gao Q, Gao X, Nie F (2018) Svm based multi-label learning with missing labels for image annotation. Pattern Recognit 78:307–317
BarutçSuoglu Z, Schapire RE, Troyanskaya OG (2006) Hierarchical multi-label prediction of gene function. Bioinformatics 22(7):830–6
Liu H, Motoda H (2008) Computational methods of feature selection
Soheili M, Eftekhari Moghadam AM (2016) Feature selection in multi-label classification through mlqpfs, pp. 430–434
Li Y-F, Hu J-A, Jiang Y, Zhou Z-H (2012) Towards discovering what patterns trigger what labels, proceedings of the national conference on. Artif Intell 2:1012–1018
Chen T-t, Liu K, Ding X-m, Zou H, Shen Q, Liu Y (2015) A multi-instance multi-label learning algorithm based on feature selection, pp. 587–590
Chen Y-N, Lin H-T (2012) Feature-aware label space dimension reduction for multi-label classification. Adv Neural Inf Process Syst 2:1529–1537
Jian L, Li J, Shu K, Liu H (2016) Multi-label informed feature selection
Valadi J, Ovhal P, Rathore K (2019) A simple method of solution for multi-label feature selection, pp. 1–4
Stone JV (2004) Principal component analysis and factor analysis, pp. 129–135
Kashef S, Nezamabadi-pour H (2017) An effective method of multi-label feature selection employing evolutionary algorithms, in: 2017 2nd Conference on Swarm Intelligence and Evolutionary Computation (CSIEC), pp. 21–25
Zhang J, Luo Z, Li C, Zhou C, Li S (2019) Manifold regularized discriminative feature selection for multi-label learning. Pattern Recognit 95:136–150
Weng W, Chen Y-N, Chen C-L, Wu S, Liu J (2020) Non-sparse label specific features selection for multi-label classification. Neurocomputing 377:85–94
Tai F, Lin H-T (2012) Multilabel classification with principal label space transformation. Neural Comput 24(9):2508–2542
Lin Z, Ding G, Hu M, Wang J (2014) Multi-label classification via feature-aware implicit label space encoding, in: E. P. Xing, T. Jebara (Eds.), Proceedings of the 31st International Conference on Machine Learning, Vol. 32 of Proceedings of Machine Learning Research, PMLR, Bejing, China, pp. 325–333
Kumar V, Pujari AK, Padmanabhan V, Kagita VR , Group preserving label embedding for multi-label classification, http://arxiv.org/abs/1812.09910
Bhatia K, Jain H, Kar P, Varma M, Jain P (2015) Sparse local embeddings for extreme multi-label classification. In: Cortes C, Lawrence N, Lee D, Sugiyama M, Garnett R (eds) Advances in neural information processing systems, vol 28. Curran Associates Inc, Cham
Huang J, Zhang P, Zhang H, Li G, Rui H (2020) Multi-label learning via feature and label space dimension reduction. IEEE Access 8:20289–20303
Abeo TA, Jun Shen X, Gou J, Mao Q, Bao B, Li S (2019) Dictionary-induced least squares framework for multi-view dimensionality reduction with multi-manifold embeddings. IET Comput Vis 13:97–108
Ganaa ED, Jun Shen X, Abeo TA (2021) Deflated manifold embedding pca framework via multiple instance factorings. Multim Tools Appl 80:3809–3833
Roweis ST, Saul LK (2000) Nonlinear dimensionality reduction by locally linear embedding. Science 290:5500
Belkin M, Niyogi P (2003) Laplacian eigenmaps for dimensionality reduction and data representation. Neural Comput 15(6):1373–1396
He X, Niyogi P (2003) Locality preserving projections, in: NIPS
Zhu Y, Kwok JT, Zhou Z-H (2018) Multi-label learning with global and local label correlation. IEEE Trans Knowledge Data Eng 30:1081–1094
Huang R, Jiang W, Sun G (2018) Manifold-based constraint laplacian score for multi-label feature selection. Pattern Recognit Lett 112:346–352
Huang R, Wu Z (2021) Multi-label feature selection via manifold regularization and dependence maximization. Pattern Recognit 120:108149
Lin Z, Chen M, Ma Y, The augmented lagrange multiplier method for exact recovery of corrupted low-rank matrices, arXiv:1009.5055
Liu W, Xu D, Tsang IW-H, Zhang W (2019) Metric learning for multi-output tasks. IEEE Trans Pattern Anal Mach Intell 41:408–422
Tsoumakas G, Xioufis ES, Vilcek J, Vlahavas IP (2011) Mulan: a java library for multi-label learning. J Mach Learn Res 12:2411–2414
Tsoumakas G, Katakis I, Vlahavas I (2011) Random k-labelsets for multilabel classification. IEEE Trans Knowl Data Eng 23(7):1079–1089
Acknowledgements
This research was funded in part by Primary Research & Development Plan of Jiangsu Province (BE2018627)
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Liu, ZF., Cai, RH., Abeo, T.A. et al. Geometrically Preserved Dual Projections Learning for Multi-label Classification. Neural Process Lett 55, 7369–7392 (2023). https://doi.org/10.1007/s11063-023-11265-6
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11063-023-11265-6