Label-Specific Feature Augmentation for Long-Tailed Multi-Label Text Classification

被引:0
|
作者
Xu, Pengyu [1 ]
Xiao, Lin [1 ]
Liu, Bing [1 ]
Lu, Sijin [1 ]
Jing, Liping [1 ]
Yu, Jian [1 ]
机构
[1] Beijing Jiaotong Univ, Beijing Key Lab Traff Data Anal & Min, Beijing, Peoples R China
来源
THIRTY-SEVENTH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE, VOL 37 NO 9 | 2023年
基金
北京市自然科学基金; 中国国家自然科学基金;
关键词
D O I
暂无
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Multi-label text classification (MLTC) involves tagging a document with its most relevant subset of labels from a label set. In real applications, labels usually follow a long-tailed distribution, where most labels (called as tail-label) only contain a small number of documents and limit the performance of MLTC. To facilitate this low-resource problem, researchers introduced a simple but effective strategy, data augmentation (DA). However, most existing DA approaches struggle in multi-label settings. The main reason is that the augmented documents for one label may inevitably influence the other co-occurring labels and further exaggerate the long-tailed problem. To mitigate this issue, we propose a new pair-level augmentation framework for MLTC, called Label-Specific Feature Augmentation (LSFA), which merely augments positive feature-label pairs for the tail-labels. LSFA contains two main parts. The first is for label-specific document representation learning in the high-level latent space, the second is for augmenting tail-label features in latent space by transferring the documents second-order statistics (intra-class semantic variations) from head-labels to tail-labels. At last, we design a new loss function for adjusting classifiers based on augmented datasets. The whole learning procedure can be effectively trained. Comprehensive experiments on benchmark datasets have shown that the proposed LSFA outperforms the state-of-the-art counterparts.
引用
收藏
页码:10602 / 10610
页数:9
相关论文
共 50 条
  • [21] Partial multi-label learning via label-specific feature corrections
    Hang, Jun-Yi
    Zhang, Min-Ling
    SCIENCE CHINA-INFORMATION SCIENCES, 2025, 68 (03)
  • [22] Probability Guided Loss for Long-Tailed Multi-Label Image Classification
    Lin, Dekun
    THIRTY-SEVENTH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE, VOL 37 NO 2, 2023, : 1577 - 1585
  • [23] Distributionally Robust Loss for Long-Tailed Multi-label Image Classification
    Lin, Dekun
    Peng, Tailai
    Chen, Rui
    Xie, Xinran
    Qin, Xiaolin
    Cui, Zhe
    COMPUTER VISION - ECCV 2024, PT XXXIII, 2025, 15091 : 417 - 433
  • [24] Effect of Stage Training for Long-Tailed Multi-Label Image Classification
    Yamagishi, Yosuke
    Hanaoka, Shohei
    2023 IEEE/CVF INTERNATIONAL CONFERENCE ON COMPUTER VISION WORKSHOPS, ICCVW, 2023, : 2713 - 2720
  • [25] Learning Label-Specific Multiple Local Metrics for Multi-Label Classification
    Mao, Jun-Xiang
    Hang, Jun-Yi
    Zhang, Min-Ling
    PROCEEDINGS OF THE THIRTY-THIRD INTERNATIONAL JOINT CONFERENCE ON ARTIFICIAL INTELLIGENCE, IJCAI 2024, 2024, : 4742 - 4750
  • [26] Learning Label-Specific Features for Multi-Label Classification with Missing Labels
    Huang, Jun
    Qin, Feng
    Zheng, Xiao
    Cheng, Zekai
    Yuan, Zhixiang
    Zhang, Weigang
    2018 IEEE FOURTH INTERNATIONAL CONFERENCE ON MULTIMEDIA BIG DATA (BIGMM), 2018,
  • [27] Multi-label feature selection based on stable label relevance and label-specific features
    Yang, Yong
    Chen, Hongmei
    Mi, Yong
    Luo, Chuan
    Horng, Shi-Jinn
    Li, Tianrui
    INFORMATION SCIENCES, 2023, 648
  • [28] Label-specific feature selection and two-level label recovery for multi-label classification with missing labels
    Ma, Jianghong
    Chow, Tommy W. S.
    NEURAL NETWORKS, 2019, 118 : 110 - 126
  • [29] Label-specific multi-label text classification based on dynamic graph convolutional networksLabel-specific multi-label text classification...Y. Yan et al.
    Yaoyao Yan
    Fang‘ai Liu
    Kenan Liu
    Weizhi Xu
    Xuqiang Zhuang
    Soft Computing, 2025, 29 (3) : 1897 - 1909
  • [30] Learning label-specific features with global and local label correlation for multi-label classification
    Weng, Wei
    Wei, Bowen
    Ke, Wen
    Fan, Yuling
    Wang, Jinbo
    Li, Yuwen
    APPLIED INTELLIGENCE, 2023, 53 (03) : 3017 - 3033