A journal of IEEE and CAA , publishes high-quality papers in English on original theoretical/experimental research and development in all areas of automation
Advanced Search
Volume 13 Issue 8
Aug.  2026

IEEE/CAA Journal of Automatica Sinica

  • JCR Impact Factor: 18.3, Top 1 (SCI Q1)
    CiteScore: 28.2, Top 1% (Q1)
    Google Scholar h5-index: 95， TOP 5
Turn off MathJax
Article Contents
Article Navigation >  IEEE/CAA Journal of Automatica Sinica  > 2026  >  13(8): 1842-1853
Q. Wu, S. Wang, M. C. Zhou, C. Chen, and Y. Gong, “Multi-label causal feature selection based on unary approximate Markov blankets,” IEEE/CAA J. Autom. Sinica, vol. 13, no. 8, pp. 1842–1853, Aug. 2026. doi: 10.1109/JAS.2026.126167
Citation: Q. Wu, S. Wang, M. C. Zhou, C. Chen, and Y. Gong, “Multi-label causal feature selection based on unary approximate Markov blankets,” IEEE/CAA J. Autom. Sinica, vol. 13, no. 8, pp. 1842–1853, Aug. 2026. doi: 10.1109/JAS.2026.126167
shu

Multi-Label Causal Feature Selection Based on Unary Approximate Markov Blankets

doi: 10.1109/JAS.2026.126167
Funds:  This work was supported in part by the National Natural Science Foundation of China (62322601, 62172065) and the National Science and Technology Major Project (2025ZD1700805)
More Information
    Abstract
  • Multi-label feature selection (FS) plays a vital role in multi-label learning as properly selected features can be used to substantively improve classification performance and reduce training time of a classifier. Existing multi-label FS methods primarily rely on the correlations among variables but neglect causality, and consequently lack interpretability. Although causal FS techniques based on Markov blankets have been widely investigated for single-label learning, its exploration for multi-label learning is rather limited because of the more complex causal relationships in multi-label data. In this paper, we present a multi-label causal FS method that leverages the introduced concept of unary approximate Markov blankets to identify causal structure of labels. Moreover, it combines the label-label, feature-feature and label-feature relationships in multi-label datasets and restores the features that are omitted due to equivalent information among features and labels. We conduct experiments on a variety of multi-label datasets and compare our proposed method with the state-of-the-art algorithms. The results show that our approach achieves significantly better performance in terms of a number of different metrics than them, thus greatly advancing the field of multi-label FS.

     

    • FullText(HTML)
  • loading

    • 1 https://mulan.sourceforge.net/datasets-mlc.html

    References(50)
  • [1]
    W. Liu, H. Wang, X. Shen, and I. W. Tsang, “The emerging trends of multi-label learning,” IEEE Trans. Pattern Anal. Mach. Intell., vol. 44, no. 11, pp. 7955–7974, Nov. 2022. doi: 10.1109/TPAMI.2021.3119334
    [2]
    Y. Lin, Z. Yu, K. Yang, Z. Fan, and C. L. P. Chen, “Boosting adaptive weighted broad learning system for multi-label learning,” IEEE/CAA J. Autom. Sinica, vol. 11, no. 11, pp. 2204–2219, Nov. 2024. doi: 10.1109/JAS.2024.124557
    [3]
    G. Lyu, S. Feng, Y. Jin, T. Wang, C. Lang, and Y. Li, “Prior knowledge regularized self-representation model for partial multilabel learning,” IEEE Trans. Cybern., vol. 53, no. 3, pp. 1618–1628, Mar. 2023. doi: 10.1109/TCYB.2021.3107422
    [4]
    S. Han, K. Zhu, M. C. Zhou, H. Alhumade, and A. Abusorrah, “Locating multiple equivalent feature subsets in feature selection for imbalanced classification,” IEEE Trans. Knowl. Data Eng., vol. 35, no. 9, pp. 9195–9209, Sep. 2023. doi: 10.1109/TKDE.2022.3222047
    [5]
    Z. Wang, S. Gao, Z. Lei, and M. Omura, “An information-based elite-guided evolutionary algorithm for multi-objective feature selection,” IEEE/CAA J. Autom. Sinica, vol. 11, no. 1, pp. 264–266, Jan. 2024. doi: 10.1109/JAS.2023.123810
    [6]
    Q. Han, Z. Zhao, L. Hu, and W. Gao, “Enhanced multi-label feature selection considering label-specific relevant information,” Expert Syst. Appl., vol. 264, Art. no. 125819, Mar. 2025. doi: 10.1016/j.eswa.2024.125819
    [7]
    J. Zhang, Y. Lin, M. Jiang, S. Li, Y. Tang, and K. C. Tan, “Multi-label feature selection via global relevance and redundancy optimization,” in Proc. 29th Int. Joint Conf. Artificial Intelligence, Yokohama, Japan, 2020, pp. 2512−2518.
    [8]
    J. Zhang, Z. Luo, C. Li, C. Zhou, and S. Li, “Manifold regularized discriminative feature selection for multi-label learning,” Pattern Recognit., vol. 95, pp. 136–150, Nov. 2019. doi: 10.1016/j.patcog.2019.06.003
    [9]
    J. Lee and D.-W. Kim, “Fast multi-label feature selection based on information-theoretic feature ranking,” Pattern Recognit., vol. 48, no. 9, pp. 2761–2771, Sep. 2015. doi: 10.1016/j.patcog.2015.04.009
    [10]
    J. Lee and D.-W. Kim, “SCLS: Multi-label feature selection based on scalable criterion for large label set,” Pattern Recognit., vol. 66, pp. 342–352, Jun. 2017. doi: 10.1016/j.patcog.2017.01.014
    [11]
    S. Kashef and H. Nezamabadi-Pour, “A label-specific multi-label feature selection algorithm based on the Pareto dominance concept,” Pattern Recognit., vol. 88, pp. 654–667, Apr. 2019. doi: 10.1016/j.patcog.2018.12.020
    [12]
    L. Jian, J. Li, K. Shu, and H. Liu, “Multi-label informed feature selection,” in Proc. 25th Int. Joint Conf. Artificial Intelligence, New York, USA, 2016, pp. 1627−1633.
    [13]
    K. Yu, X. Guo, L. Liu, J. Li, H. Wang, Z. Ling, and X. Wu, “Causality-based feature selection: Methods and evaluations,” ACM Comput. Surv., vol. 53, no. 5, Art. no. 111, Sep. 2020.
    [14]
    K. Yu, M. Cai, X. Wu, L. Liu, and J. Li, “Multilabel feature selection: A local causal structure learning approach,” IEEE Trans. Neural Networks Learn. Syst., vol. 34, no. 6, pp. 3044–3057, Jun. 2023. doi: 10.1109/TNNLS.2021.3111288
    [15]
    X. Wu, B. Jiang, K. Yu, H. Chen, and C. Miao, “Multi-label causal feature selection,” in Proc. 34th AAAI Conf. Artificial Intelligence, New York, USA, 2020, pp. 6430−6437.
    [16]
    X. Guo, K. Yu, F. Cao, P. Li, and H. Wang, “Error-aware Markov blanket learning for causal feature selection,” Inf. Sci., vol. 589, pp. 849–877, Apr. 2022. doi: 10.1016/j.ins.2021.12.118
    [17]
    J.-P. Pellet and A. Elisseeff, “Using Markov blankets for causal structure learning,” J. Mach. Learn. Res., vol. 9, pp. 1295–1342, Jun. 2008.
    [18]
    I. Tsamardinos, C. F. Aliferis, and A. Statnikov, “Time and sample efficient discovery of Markov blankets and direct causal relations,” in Proc. 9th ACM SIGKDD Int. Conf. Knowledge Discovery and Data Mining, Washington, USA, 2003, pp. 673−678.
    [19]
    C. F. Aliferis, I. Tsamardinos, and A. R. Statnikov, “HITON: A novel Markov blanket algorithm for optimal variable selection,” in Proc. American Medical Informatics Association Annu. Symp., Washington, USA, 2003, pp. 21−25.
    [20]
    Z. Ling, Y. Li, Y. Zhang, K. Yu, P. Zhou, B. Li, and X. Wu, “A light causal feature selection approach to high-dimensional data,” IEEE Trans. Knowl. Data Eng., vol. 35, no. 8, pp. 7639–7650, Aug. 2023. doi: 10.1109/tkde.2022.3218786
    [21]
    A. Statnikov, J. Lemeir, and C. F. Aliferis, “Algorithms for discovery of multiple Markov boundaries,” J. Mach. Learn. Res., vol. 14, no. 1, pp. 499–566, Feb. 2013.
    [22]
    G. Chandrashekar and F. Sahin, “A survey on feature selection methods,” Comput. Electr. Eng., vol. 40, no. 1, pp. 16–28, Jan. 2014. doi: 10.1016/j.compeleceng.2013.11.024
    [23]
    J. Li, K. Cheng, S. Wang, F. Morstatter, R. P. Trevino, J. Tang, and H. Liu, “Feature selection: A data perspective,” ACM Comput. Surv., vol. 50, no. 6, Art. no. 94, Nov. 2018.
    [24]
    S. Kashef, H. Nezamabadi-Pour, and B. Nikpour, “Multilabel feature selection: A comprehensive review and guiding experiments,” WIREs Data Min. Knowl. Discovery, vol. 8, no. 2, Art. no. e1240, Mar.–Apr. 2018. doi: 10.1002/widm.1240
    [25]
    H. Liu, M. C. Zhou, and Q. Liu, “An embedded feature selection method for imbalanced data classification,” IEEE/CAA J. Autom. Sinica, vol. 6, no. 3, pp. 703–715, May 2019. doi: 10.1109/jas.2019.1911447
    [26]
    L. Li, H. Liu, Z. Ma, Y. Mo, Z. Duan, J. Zhou, and J. Zhao, “Multi-label feature selection via information gain,” in Proc. 10th Int. Conf. Advanced Data Mining and Applications, Guilin, China, 2014, pp. 345−355.
    [27]
    D. Paul, R. Kumar, S. Saha, and J. Mathew, “Multi-objective cuckoo search-based streaming feature selection for multi-label dataset,” ACM Trans. Knowl. Discovery Data, vol. 15, no. 6, Art. no. 93, Jun. 2021.
    [28]
    R. Alhutaish and N. Omar, “Feature selection for multi-label document based on wrapper approach through class association rules,” Int. J. Adv. Sci., Eng. Inf. Technol., vol. 7, no. 2, pp. 642–649, Apr. 2017.
    [29]
    A. A. Bidgoli, H. Ebrahimpour-Komleh, and S. Rahnamayan, “Reference-point-based multi-objective optimization algorithm with opposition-based voting scheme for multi-label feature selection,” Inf. Sci., vol. 547, pp. 1–17, Feb. 2021. doi: 10.1016/j.ins.2020.08.004
    [30]
    J. Dai, W. Chen, and Y. Qian, “Multi-label feature selection with missing features via implicit label replenishment and positive correlation feature recovery,” IEEE Trans. Knowl. Data Eng., vol. 37, no. 4, pp. 2042–2055, Apr. 2025. doi: 10.1109/TKDE.2025.3536080
    [31]
    X. Wu, B. Jiang, Y. Zhong, and H. Chen, “Tolerant Markov boundary discovery for feature selection,” in Proc. 29th ACM Int. Conf. Information & Knowledge Management, Ireland, 2020, pp. 2261−2264.
    [32]
    X. Wu, B. Jiang, X. Wang, T. Ban, and H. Chen, “Feature selection in the data stream based on incremental Markov boundary learning,” IEEE Trans. Neural Networks Learn. Syst., vol. 34, no. 10, pp. 6740–6754, Oct. 2023. doi: 10.1109/TNNLS.2023.3249767
    [33]
    X. Yang, Y. Wang, Y. Ou, and Y. Tong, “Three-fast-inter incremental association Markov blanket learning algorithm,” Pattern Recognit. Lett., vol. 122, pp. 73–78, May 2019. doi: 10.1016/j.patrec.2019.02.002
    [34]
    I. Tsamardinos, G. Borboudakis, P. Katsogridakis, P. Pratikakis, and V. Christophides, “A greedy feature selection algorithm for big data of high dimensionality,” Mach. Learn., vol. 108, no. 2, pp. 149–202, Feb. 2019. doi: 10.1007/s10994-018-5748-7
    [35]
    G. Borboudakis and I. Tsamardinos, “Forward-backward selection with early dropping,” J. Mach. Learn. Res., vol. 20, no. 1, pp. 276–314, Jan. 2019.
    [36]
    X. Wu, B. Jiang, K. Yu, C. Miao, and H. Chen, “Accurate Markov boundary discovery for causal feature selection,” IEEE Trans. Cybern., vol. 50, no. 12, pp. 4983–4996, Dec. 2020. doi: 10.1109/TCYB.2019.2940509
    [37]
    X. Wu, B. Jiang, T. Wu, and H. Chen, “Practical Markov boundary learning without strong assumptions,” in Proc. 37th AAAI Conf. Artificial Intelligence, Washington, USA, 2023, pp. 10388−10398.
    [38]
    Z. Ling, J. Wu, Y. Zhang, P. Zhou, X. Wu, K. Yu, and X. Wu, “Label-aware causal feature selection,” IEEE Trans. Knowl. Data Eng., vol. 37, no. 3, pp. 1268–1281, Mar. 2025. doi: 10.1109/TKDE.2024.3522580
    [39]
    X.-Q. Liu and X.-S. Liu, “Markov blanket and Markov boundary of multiple variables,” J. Mach. Learn. Res., vol. 19, no. 1, pp. 1658–1707, Jan. 2018.
    [40]
    X. Wu, B. Jiang, Y. Zhong, and H. Chen, “Multi-target Markov boundary discovery: Theory, algorithm, and application,” IEEE Trans. Pattern Anal. Mach. Intell., vol. 45, no. 4, pp. 4964–4980, Apr. 2023. doi: 10.1109/TPAMI.2022.3199784
    [41]
    L. Ma, L. Hu, Y. Li, W. Ding, and W. Gao, “MI-MCF: A mutual information-based multilabel causal feature selection,” IEEE Trans. Neural Networks Learn. Syst., vol. 36, no. 6, pp. 9864–9878, Jun. 2025. doi: 10.1109/TNNLS.2025.3556128
    [42]
    W. Gao, P. Hao, Y. Wu, and P. Zhang, “A unified low-order information-theoretic feature selection framework for multi-label learning,” Pattern Recognit., vol. 134, Art. no. 109111, Feb. 2023. doi: 10.1016/j.patcog.2022.109111
    [43]
    Y. Du, Y. Sun, and L. Tan, “Ordinal causal discovery based on Markov blankets,” Comput. Stat., vol. 40, no. 3, pp. 1311–1335, Mar. 2025. doi: 10.1007/s00180-024-01513-1
    [44]
    Z. Ling, K. Yu, Y. Zhang, L. Liu, and J. Li, “Causal learner: A toolbox for causal structure and Markov blanket learning,” Pattern Recognit. Lett., vol. 163, pp. 92–95, Nov. 2022. doi: 10.1016/j.patrec.2022.09.021
    [45]
    J. Liang, J. Wang, G. Yu, C. Domeniconi, X. Zhang, and M. Guo, “Gradient-based local causal structure learning,” IEEE Trans. Cybern., vol. 54, no. 1, pp. 486–495, Jan. 2024. doi: 10.1109/TCYB.2023.3237635
    [46]
    J. Peters, D. Janzing, and B. Schölkopf, Elements of Causal Inference: Foundations and Learning Algorithms. Cambridge, USA: The MIT Press, 2017.
    [47]
    T. Gao and Q. Ji, “Efficient Markov blanket discovery and its application,” IEEE Trans. Cybern., vol. 47, no. 5, pp. 1169–1179, May 2017. doi: 10.1109/TCYB.2016.2539338
    [48]
    M.-L. Zhang and Z.-H. Zhou, “ML-KNN: A lazy learning approach to multi-label learning,” Pattern Recognit., vol. 40, no. 7, pp. 2038–2048, Jul. 2007. doi: 10.1016/j.patcog.2006.12.019
    [49]
    M.-L. Zhang and Z.-H. Zhou, “A review on multi-label learning algorithms,” IEEE Trans. Knowl. Data Eng., vol. 26, no. 8, pp. 1819–1837, Aug. 2014. doi: 10.1109/TKDE.2013.39
    [50]
    J. Demšar, “Statistical comparisons of classifiers over multiple data sets,” J. Mach. Learn. Res., vol. 7, pp. 1–30, Dec. 2006.
    • Related Articles
    • Supplements(0)

Catalog

    通讯作者: 陈斌, bchen63@163.com
    • 1. 

      沈阳化工大学材料科学与工程学院 沈阳 110142

    1. 本站搜索
    2. 百度学术搜索
    3. 万方数据库搜索
    4. CNKI搜索

    Figures(8)  / Tables(11)

    Get Citation
    PDF
    XML

    Article Metrics

    Article views (22) PDF downloads(2) Cited by()

    /

    DownLoad:  Full-Size Img  PowerPoint
    Return
    Return