Abstract
Multi-label learning is more complicated than single-label learning since the semantics of the instances are usually overlapped and not identical. The effectiveness of many algorithms often fails when the correlations in the feature and label space are not fully exploited. To this end, we propose a novel non-negative matrix factorization (NMF) based modeling and training algorithm that learns from both the adjacencies of the instances and the labels of the training set. In the modeling process, a set of generators are constructed, and the associations among generators, instances, and labels are set up, with which the label prediction is conducted. In the training process, the parameters involved in the process of modeling are determined. Specifically, an NMF based algorithm is proposed to determine the associations between generators and instances, and a non-negative least square optimization algorithm is applied to determine the associations between generators and labels. The proposed algorithm fully takes the advantage of smoothness assumption, so that the labels are properly propagated. The experiments were carried out on six set of benchmarks. The results demonstrate the effectiveness of the proposed algorithms.
Similar content being viewed by others
References
Zhang M L, Zhou Z H. A review on multi-label learning algorithms. IEEE Transactions on Knowledge and Data Engineering, 2014, 26(8): 1819–1837
Zhang M L, Zhou Z H. Milti-label neural networks with applications to functional genomics and text categorization. IEEE Transactions on Knowledge and Data Engineering, 2006, 18(10): 1338–1351
Lo H, Wang J, Wang H, Lin S. Cost sensitive multi-label learning for audio tag annotation and retrieval. IEEE Transactions on Multimedia, 2011, 13(3): 518–529
Sanden C, Zhang J. Enhancing multi-label music genre classification through ensemble techniques. In: Proceedings of the 34th International ACMSIGIR Conference on Research and Development in Information Retrieval. 2011, 705–714
Tang L, Rajan S, Narayanan V. Large scale multi-label classification via metalabeler. In: Proceedings of the 19th International Conference on World Wide Web. 2009, 211–220
Gopal S, Yang Y. Multi-label classification with meta-level features. In: Proceedings of the 33rd International ACM SIGIR Conference on Research & Development in Information Retrieval. 2010, 315–322
Zhu X, Ghahramani Z. Learning from labeled and unlabeled data with label propagation. Technical Report, 2002
Read J, Martino L, Olmos P M, Luengo D. Scalable multi-output label prediction: from classifier chains to classifier trellises. Pattern Recognition, 2015, 48(6): 2096–2109
Madjarov G, Gjorgjevikj D, Dzeroski S. Two stage architecture for multi-label learning. Pattern Recognition, 2012, 45(3): 1019–1034
Teisseyre P. Feature ranking for multi-label classification usingMarkov networks. Neurocomputing, 2015, 205: 439–454
Lee J, Kim D W. Memetic feature selection algorithm for multi-label classification. Information Sciences, 2015, 293: 80–96
Zhu X, Lafferty J, Rosenfeld R. Semi-supervised learning with graphs. Carnegie Mellon University, Doctor Thesis, 2005
Chapelle O, Schlkopf B, Zien A. Semi-Supervised Learning. MA: The MIT Press, 2006
Hou P, Geng X, Zhang M L. Multi-label manifold learning. In: Proceedings of the 30th AAAI Conference on Artificial Intelligence. 2016, 1680–1686
Gao N N, Huang S J, Chen S C. Multi-label active learning by model guided distribution matching. Frontiers of Computer Science, 2016, 10(5): 845–855
Kong X, Ng M K, Zhou Z H. Transductive multi-label learning via label set propagation. IEEE Transactions on Knowledge and Data Engineering, 2013, 25(3): 704–719
Huang S J, Yu Y, Zhou Z H. Multi-label hypothesis reuse. In: Proceedings of the 18th ACM SIGKDD Conference on Knowledge Discovery and Data Mining. 2012, 525–533
Huang S J, Zhou Z H. Multi-label learning by exploiting label correlations locally. In: Proceedings of the 26th AAAI Conference on Artificial Intelligence. 2012, 949–955
Lo H, Lin S,Wang H. Generalized k-label sets ensemble for multi-label and cost-sensitive classification. IEEE Transactions on Knowledge and Data Engineering, 2014, 26(7): 1679–1691
Lee J, Kim K, Kim N, Lee J H. An approach for multi-label classification by directed acyclic graph with label correlation maximization. Information Sciences, 2016, 351: 101–114
Li P, Li H, Wu M. Multi-label ensemble based on variable pairwise constraint projection. Information Sciences, 2013, 222: 269–281
Wang B, Tsotsos J. Dynamic label propagation for semi-supervised multi-class multi-label classification. Pattern Recognition, 2016, 52: 75–84
Wang S, Wang J, Wang Z, Ji Q. Enhancing multi-label classification by modeling dependencies among labels. Pattern Recognition, 2014, 47(10): 3405–3413
Pillai I, Fumera G, Roli F. Multi-label classification with a reject option. Pattern Recognition, 2013, 46(8): 2256–2266
Sun F, Tang J, Li H, Qi G, Huang T S. Multi-label image categorization with sparse factor representation. IEEE Transactions on Image Processing, 2014, 23(3): 1028–1037
Zhang Y, Zhou Z H. Multi-label dimensionality reduction via dependence maximization. ACM Transactions on Knowledge Discovery from Data, 2010, 4(3): 14
Zhang M L, Zhang K. Multi-label learning by exploiting label dependency. In: Proceedings of the 16th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. 2010, 999–1007
Zhang M L, Wu L. Lift: multi-label learning with label specific features. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2015, 37(1): 107–120
Triguero I, Vens C. Labelling strategies for hierarchical multi-label classification techniques. Pattern Recognition, 2016, 56: 170–183
Quevedo J R, Luaces O, Bahamonde A. Multilabel classifiers with a probabilistic thresholding strategy. Pattern Recognition, 2012, 45(2): 876–883
Zhang N, Ding S, Zhang J. Multi layer ELM-RBF for multi-label learning. Applied Soft Computing, 2016, 43: 535–545
Huang Y, Wang W, Wang L. Unconstrained multimodal multi-label learning. IEEE Transactions on Multimedia, 2015, 17(11): 1923–1935
Xu J, Jagadeesh V, Manjunath B S. Multi-label learning with fused multimodal Bi-relational graph. IEEE Transactions on Multimedia, 2014, 16(2): 403–412
Gao W, Zhou Z. On the consistency of multi-label learning. Artificial Intelligence, 2013, 199(1): 22–44
Madjarov G, Kocev D, Gjorgjevikj D, Dzeroski S. An extensive experimental comparison of methods for multi-label learning. Pattern Recognition, 2012, 45(9): 3084–3104
Lee D D, Seung H S. Learning the parts of objects by non-negative matrix factorization. Nature, 1999, 401(6755): 788–791
Lawson C L, Hanson R J. Solving Least Squares Problems. New Jersey: Prentice-Hall, Inc., 1974
Joachims T. Transductive inference for text classification using support vector machines. In: Proceedings of the 16th International Conference on Machine Learning. 1999, 200–209
Vapnik V N. Statistical Learning Theory. New York: Wiley, 1998
Belkin M, Niyogi P, Sindhwani V. Manifold regularization: a geometric framework for learning from examples. Journal of Machine Learning Research, 2006, 7: 2399–2434
Xu M, Jin R, Zhou Z H. Speedup matrix completion with side information: application to multi-label learning. In: Proceedings of the 27th Annual Conference on Neural Information Processing Systems. 2013, 2301–2309
Read J, Pfahringer B, Holmes G, Frank E. Classifier chains for multilabel classification. Machine Learning, 2011, 85(3): 333–359
Wang J, Zhao Y, Wu X, Hua X S. A transductive multi-label learning approach for video concept detection. Pattern Recognition, 2011, 44(10–11): 2274–2286
Furnkranz J, Hullermeier E, Mencia E L, Brinker K. Multilabel classification via calibrated label ranking. Machine Learning, 2008, 73(2): 133–153
Mencia E L, Park S H, Furnkranz J. Efficient voting prediction for pairwise multi-label classification. Neurocomputing, 2010, 73(7–9): 1164–1176
Acknowledgements
The authors are grateful to the support of the National Natural Science Foundation of China (Grant Nos. 61402076, 61572104, 61103146), the Fundamental Research Funds for the Central Universities (DUT17JC04), and the Project of the Key Laboratory of Symbolic Computation and Knowledge Engineering ofMinistry of Education, Jilin University (93K172017K03).
Author information
Authors and Affiliations
Corresponding author
Additional information
Liang Sun received the BE degree in computer science and technology from Xidian University, China in 2003, the MS degree and PhD degree in computer application technology from Jilin University, China in 2006 and 2012, respectively. He is currently with the College of Computer Science and Technology, Dalian university of technology, China. His main research interests lie in machine learning and deep learning.
Hongwei Ge received BS and MS degrees in mathematics from Jilin University, China, and the PhD degree in computer application technology from Jilin University, China in 2006. He is currently a professor and a vice dean in the College of Computer Science and Technology, Dalian University of Technology, China. His research interests are machine learning, computational intelligence, optimization and modeling, computer vision, and deep learning.
Wenjing Kang received the BS degree from Northeast University, China in 2016. She is currently pursuing a master degree in the College of Computer Science and Technology, Dalian University of Technology, China. Her main research interests are deep learning, machine learning applications such as computer vision.
Electronic supplementary material
Rights and permissions
About this article
Cite this article
Sun, L., Ge, H. & Kang, W. Non-negative matrix factorization based modeling and training algorithm for multi-label learning. Front. Comput. Sci. 13, 1243–1254 (2019). https://doi.org/10.1007/s11704-018-7452-y
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11704-018-7452-y