Interactive Causal Correlation Space Reshape for Multi-Label Classification.

Chao Zhang; Yusheng Cheng; Yibin Wang; Yuting Xu

doi:10.9781/ijimai.2022.08.007

Authors

Chao Zhang Anqing Normal University
Yusheng Cheng Anqing Normal University
Yibin Wang Anqing Normal University
Yuting Xu Anqing Normal University

DOI:

https://doi.org/10.9781/ijimai.2022.08.007

Keywords:

Conditional Probability, Interactive Causal Inference, Label Co-Occurrence, Label Space Reshape, Multi-Label Classification

Supporting Agencies

This work was supported by the National Natural Science Foundation of Anhui under Grant 2108085MF216 and Key Laboratory of Data Science and Intelligence Application, Fujian Province University (NO. D202005). Key Laboratory of Intelligent Computing & Signal Processing, Ministry of Education (Anhui University) (No.2020A003) and Anqing Normal University Graduate Innovation Fund (No.2021yjsXSCX017).

Abstract

Most existing multi-label classification models focus on distance metrics and feature spare strategies to extract specific features of labels. Those models use the cosine similarity to construct the label correlation matrix to constraint solution space, and then mine the latent semantic information of the label space. However, the label correlation matrix is usually directly added to the model, which ignores the interactive causality of the correlation between the labels. Considering the label-specific features based on the distance method merely may have the problem of distance measurement failure in the high-dimensional space, while based on the sparse weight matrix method may cause the problem that parameter is dependent on manual selection. Eventually, this leads to poor classifier performance. In addition, it is considered that logical labels cannot describe the importance of different labels and cannot fully express semantic information. Based on these, we propose an Interactive Causal Correlation Space Reshape for Multi-Label Classification (CCSRMC) algorithm. Firstly, the algorithm constructs the label propagation matrix using characteristic that similar instances can be linearly represented by each other. Secondly, label co-occurrence matrix is constructed by combining the conditional probability test method, which is based on the label propagation reshaping the label space to rich label semantics. Then the label co-occurrence matrix combines with the label correlation matrix to construct the label interactive causal correlation matrix to perform multi-label classification learning on the obtained numerical label matrix. Finally, the algorithm in this paper is compared with multiple advanced algorithms on multiple benchmark multi-label datasets. The results show that considering the interactive causal label correlation can reduce the redundant information in the model and improve the performance of the multi-label classifier.

Downloads

Download data is not yet available.

References

Y. B. Wang, W. J. Zheng, Y. S. Cheng, T. C. Cao, “Multi-label classification algorithm based on PLSA learning probability distribution semantic information”, Journal of Nanjing University (Natural Science), vol. 57, no. 1, pp. 75-89, 2021.

J. Huang, F. Qin, X. Zheng, “Learning label-specific features for multilabel classification”, Information Sciences: An International Journal, vol. 492, no. 18, pp. 124-146, 2019.

H. R. Han, M. X. Huang, Y. Zhang, X. G. Yang, W. G. Feng, “Multi-label learning with label specific features using correlation information”, IEEE Access, vol. 19, no. 7, pp. 11474–11484, 2019.

M. L. Zhang, L. Wu, “Multi-label learning with label-specific features”, IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 37, no. 1, pp. 107–120, 2015.

J. Huang, F. Qin, X. Zheng, “Improving multi-label classification with missing labels by learning label-specific features”, Information Sciences: An International Journal, vol. 492, pp. 124–146, 2019.

Y. S. Cheng, K. Qian, Y. B. Wang, D. W. Zhao, “Multi-label lazy learning approach based on firefly method”, Journal of Computer Applications, vol. 39, no. 5, pp. 1305–1311, 2019.

G. M. Zhang, B. Y. Long, J. X. Zeng, J. Y. Huang, “Zero-shot attribute recognition based on de-redundancy features and semantic relationship constraint”, Pattern Recognition and Artificial Intelligence, vol. 34, no. 9, pp. 809-823, 2021.

J. C. Bao, Y. B. Wang, Y. S. Cheng, “Asymmetry label correlation for multi-label learning”, Applied Intelligence. Prepublish (2021), doi:10. 1007/S10489-021-02725-4.

D. D. Lai, Z. H. Luo, Y. L. Ma, “Label order optimization method of classifier chains based on co-occurrence analysis”, Systems Engineering and Electronics, vol. 43, no. 9, pp. 2526-2534, 2021.

C. X. Yan, S. G. Zhou, “Effective and scalable causal partitioning based on low-order conditional independent tests”, Neurocomputing, vol. 389, no. 14, pp. 146-154, 2020.

X. S. Yin, T. Shu, Q. Huang, “Semi-supervised fuzzy clustering with metric learning and entropy regularization”, Knowledge-based systems, vol. 35, pp. 304-311, 2012.

W. B. Qian, Y. S. Xiong, J. Yang, W. H. Shu, “Feature selection for label distribution learning via feature similarity and label correlation” Information Sciences, vol. 582, pp. 38–59, 2022.

Y. S. Cheng, C. Zhang, S. F. Pang, “Multi label space reshape for semantic rich label specific features learning”, International Journal of Machine Learning and Cybernetics, vol. 13, pp. 1005–1019, 2022, doi:10.1007/s13042-021-01432-3.

A. Kale, Y. F. Wu, J. Hullman, “Causal support: modeling causal inferences with visualizations”, IEEE transactions on visualization and computer graphics, 2021, doi:10.1109/TVCG.2021.3114824.

X. J. Song, A. Taamouti, “Measuring granger causal ity in quantiles”, Journal of Business & Economic Statistics, vol. 39, no. 4, pp. 1-48, 2021.

De La Pava Panche Iván, Álvarez Meza Andrés, Herrera Gómez Paula Marcela, Cárdenas Peña David, Ríos Patiño Jorge Iván, Orozco Gutiérrez Álvaro, “Kernel-based phase transfer entropy with enhanced feature relevance analysis for brain computer interfaces”, Applied Sciences, vol. 11, no. 15, pp. 6689-6689, 2021.

Z. C. Sha, Z. M. Liu, C. Ma, J. Chen, “Feature selection for multi-label classification by maximizing full-dimensional conditional mutual information”, Applied Intelligence, vol. 51, no. 1, pp. 326-340, 2020.

X. Y. Li, H. L. Wang, B. Y. Wu, “A stable and efficient technique for linear boundary value problems by applying kernel functions”, Applied Numerical Mathematics, vol. 172, no. 1, pp. 206-214, 2022.

G. J. Székely, M. L. Rizzo, N. K. Bakirov, “Measuring and testing dependence by correlation of distances”, The Annals of Statistics, vol. 35, pp. 2769–2794, 2007

R. C. Cai, Y. M. Bai, J. Qiao, Z. F. Hao, “Causal inference method based on confounder hidden compact representation model”, Journal of Computer Applications, 2021, doi: 10.11772/j.issn.1001-9081.2020122066.

Z. H. Zhou, M. L. Zhang, “Multi-label learning”, Encyclopedia of Machine Learning and Data Mining, Berlin, 2016, pp. 875–881, Springer.

Y. B. Wang, W. J. Zheng, Y. S. Cheng, D. W. Zhao, “Joint label completion and label-specific features for multi-label learning algorithm”, Soft Computing, vol. 24, pp. 6553–6569, 2020.

A. Beck, M. Teboulle, “Fast gradient-based algorithms for constrained total variation image denoising and deblurring problems”, IEEE transactions on image processing, vol. 18, no. 11, pp. 2419–2434, 2009.

G. C. Liu, Z. C. Lin, Y. Yu, “Robust subspace segmentation by low-rank representation”, Proceeding Twenty-Seventh International Conference on Machine Learning, 2010, pp. 663–670.

Z. C. Lin, A. Ganesh, J. Wright, L. Q. Wu, M. Chen, Y. Ma, “Fast convex optimization algorithms for exact recovery of a corrupted low-rank matrix”, UIUC Technical Report 2009, vol. 09, pp. 2214.

Z. H. Zhou, M. L. Zhang, “A Review on multi-label learning algorithms”, IEEE Transactions on Knowledge and Data Engineering, vol. 26, no. 8, pp. 1819–1837, 2014.

W. Weng, Y. J. Lin, Y. W. Li, “Online multi-label streaming feature selection based on neighborhood rough set”, Pattern Recognition: The Journal of the Pattern Recognition Society, vol. 84, no. 1, pp. 273–287, 2018.

J. Demiar, D. Schuurmans, “Statistical comparisons of classifiers over multiple datasets”, Journal of Machine Learning Research, vol. 7, no. 1, pp. 1–30, 2006.