Improve label embedding quality through global sensitive GAT for hierarchical text classification

被引:7
作者
Liu, Hankai [1 ]
Huang, Xianying [1 ]
Liu, Xiaoyang [1 ]
机构
[1] Chongqing Univ Technol, Coll Comp Sci & Engn, Chongqing 400054, Peoples R China
基金
中国国家自然科学基金;
关键词
Hierarchical text classification; Long-tailed distribution; Structure similarity; Density; Rebalance loss;
D O I
10.1016/j.eswa.2023.122267
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Hierarchical text classification aims to assign text to multiple labels in a label set stored in a tree structure. The current algorithms mainly introduce the priori information of the label hierarchy, but the implicit correlation between labels in the hierarchy is rarely applied. At the same time, we also found that the inherent class imbalance of chain labels will also lead to poor classification effects of lower-level labels through a large number of studies. Therefore, a label structure enhanced hierarchy aware global model (LSE-HiAGM) is proposed. Firstly, the common density coefficient of labels is defined to measure the importance of a pair of labels in the hierarchical structure. Secondly, the common density coefficient is used as the weight of the label to update the topological structure features, so that the label can be linked with all labels globally. Finally, the topological structure feature, text features, and label hierarchical features are fused to make full use of all features to improve the embedding quality of low-level labels. In addition, to alleviate the class imbalance problem, a new loss function is used to constrain the model training. The probability of the label being sampled relative to all the labels of the sample is taken as the weight of the loss function. Therefore, a small penalty is imposed on the upper label and a large penalty on the lower label. A large number of experiments on datasets such as RCV1, WOS and NYT show that LSE-HiAGM performs better than the baseline models in hierarchical text classification.
引用
收藏
页数:9
相关论文
共 24 条
  • [1] Banerjee S, 2019, 57TH ANNUAL MEETING OF THE ASSOCIATION FOR COMPUTATIONAL LINGUISTICS (ACL 2019), P6295
  • [2] Taming Pretrained Transformers for Extreme Multi-label Text Classification
    Chang, Wei-Cheng
    Yu, Hsiang-Fu
    Zhong, Kai
    Yang, Yiming
    Dhillon, Inderjit S.
    [J]. KDD '20: PROCEEDINGS OF THE 26TH ACM SIGKDD INTERNATIONAL CONFERENCE ON KNOWLEDGE DISCOVERY & DATA MINING, 2020, : 3163 - 3171
  • [3] Chen JA, 2020, 58TH ANNUAL MEETING OF THE ASSOCIATION FOR COMPUTATIONAL LINGUISTICS (ACL 2020), P2147
  • [4] Deng ZF, 2021, 2021 CONFERENCE OF THE NORTH AMERICAN CHAPTER OF THE ASSOCIATION FOR COMPUTATIONAL LINGUISTICS: HUMAN LANGUAGE TECHNOLOGIES (NAACL-HLT 2021), P3259
  • [5] Are Graph Convolutional Networks With Random Weights Feasible?
    Huang, Changqin
    Li, Ming
    Cao, Feilong
    Fujita, Hamido
    Li, Zhao
    Wu, Xindong
    [J]. IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2023, 45 (03) : 2751 - 2768
  • [6] Hierarchical Multi-label Text Classification: An Attention-based Recurrent Network Approach
    Huang, Wei
    Chen, Enhong
    Liu, Qi
    Chen, Yuying
    Huang, Zai
    Liu, Yang
    Zhao, Zhou
    Zhang, Dan
    Wang, Shijin
    [J]. PROCEEDINGS OF THE 28TH ACM INTERNATIONAL CONFERENCE ON INFORMATION & KNOWLEDGE MANAGEMENT (CIKM '19), 2019, : 1051 - 1060
  • [7] Jamal MA, 2020, PROC CVPR IEEE, P7607, DOI 10.1109/CVPR42600.2020.00763
  • [8] Johnson R., 2015, P C N AM CHAPTER ASS, V2011, P103, DOI [10.3115/v1/n15-1011, DOI 10.3115/V1/N15-1011]
  • [9] Kang Bingyi, 2019, ARXIV
  • [10] Kim Y., 2014, P 2014 C EMP METH NA