基于标签增强的细粒度文本分类

doi:10.3778/j.issn.1002-8331.2307-0129

摘要/Abstract

摘要： 文本分类是自然语言处理的一个重要分支，旨在通过训练给数据标注标签。但现有的方法仅仅考虑了标签和文本之间最浅显的语义关系，并没有考虑标签本身的额外语义信息，导致文本分类的准确率难以提升。针对此问题，提出一种基于标签增强的细粒度文本分类模型(FGTC)，它根据已知信息对标签进行解释，丰富了标签和文档之间的语义联系。此外，FGTC进一步建模标签中短语的序列关系,并采用单词级别的细粒度标签注意力方法，充分挖掘了标签的有效信息。在四个基准数据集上进行了对比实验，结果表明，该模型在文本分类任务上的准确率得到有效提升。

关键词: 文本分类, 标签增强, 细粒度标签注意力

Abstract: Text classification is an important branch of natural language processing, which aims to label data through training. However, the existing methods only consider the most obvious semantic relationship between the label and the text, and do not consider the additional semantic information of the label itself, which makes it difficult to improve the accuracy of text classification. To solve this problem, this paper proposes a label-enhanced fine-grained text classification model (FGTC), which interprets labels based on known information and enriches the semantic links between labels and documents. In addition, FGTC further models the sequence relationships of phrases in labels and adopts a fine-grained label attention method at the word level, fully mining the effective information of labels. Comparative experiments are conducted on four benchmark datasets, and the results show that the accuracy of proposed model in text classification tasks is effectively improved.

Key words: text classification, label interpretation, fine-grained label attattention

郭瑞强, 杨世龙, 贾晓文, 魏谦强. 基于标签增强的细粒度文本分类[J]. 计算机工程与应用, 2024, 60(21): 134-141.

GUO Ruiqiang, YANG Shilong, JIA Xiaowen, WEI Qianqiang. Fine-Grained Text Classification Based on Label Augmentation[J]. Computer Engineering and Applications, 2024, 60(21): 134-141.

参考文献

[1] ZHU Y, LU X, HONG J, et al. Joint dynamic topic model for recognition of lead-lag relationship in two text corpora[J]. Data Mining and Knowledge Discovery, 2022, 36(6): 2272-2298.
[2] YANG Z, YANG D, DYER C, et al. Hierarchical attention networks for document classification[C]//Proceedings of the 2016 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, 2016: 1480-1489.
[3] KHOT T, CLARK P, GUERQUIN M, et al. Qasc: a dataset for question answering via sentence composition[C]//Proceedings of the AAAI Conference on Artificial Intelligence, 2020: 8082-8090.
[4] SAXENA A, TRIPATHI A, TALUKDAR P. Improving multi-hop question answering over knowledge graphs using knowledge base embeddings[C]//Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, 2020: 4498-4507.
[5] PETRESCU A, TRUIC? C O, APOSTOL E S, et al. EDSA-ensemble: an event detection sentiment analysis ensemble architecture[J]. arXiv:2301.12805, 2023.
[6] JOHN-OTUMU A M, RAHMAN M M, NWOKONKWO O C, et al. AI-based techniques for online social media network sentiment analysis: a methodical review[J]. International Journal of Computer and Information Engineering, 2022, 16(12): 555-560.
[7] XIAO L, HUANG X, CHEN B, et al. Label-specific document representation for multi-label text classification[C]//Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP), 2019: 466-475.
[8] ZHANG X, ZHANG Q W, YAN Z, et al. Enhancing label correlation feedback in multi-label text classification via multi-task learning[J]. arXiv:2106.03103, 2021.
[9] DUAN L, YOU Q, WU X, et al. Multilabel text classification algorithm based on fusion of two-stream transformer[J]. Electronics, 2022, 11(14): 2138-2149.
[10] CHEN Y. Convolutional neural network for sentence classification[D]. Waterloo: University of Waterloo, 2015.
[11] 滕金保, 孔韦韦, 田乔鑫, 等. 基于LSTM-Attention与CNN混合模型的文本分类方法[J]. 计算机工程与应用, 2021, 57(14): 126-133.
TENG J B, KONG W W, TIAN Q X, et al. Text classification method based on LSTM-Attention and CNN hybrid model[J]. Computer Engineering and Applications, 2021, 57(14): 126-133.
[12] JOULIN A, GRAVE E, BOJANOWSKI P, et al. Bag of tricks for efficient text classification[J]. arXiv:1607.01759, 2016.
[13] YOU R, ZHANG Z, WANG Z, et al. AttentionXML: label tree?based attention?aware deep model for high?performance extreme multi-label text classification[C]//Proceedings of the 33rd International Conference on Neural Information Processing Systems, 2019.
[14] DU C, CHEN Z, FENG F, et al. Explicit interaction model towards text classification[C]//Proceedings of the AAAI Conference on Artificial Intelligence, 2019: 6359-6366.
[15] YADAV R K, JIAO L, GRANMO O C, et al. Enhancing interpretable clauses semantically using pretrained word representation[J]. arXiv:2104.06901, 2021.
[16] ZENG F, CHEN N, YANG D, et al. Simplified-boosting ensemble convolutional network for text classification[J]. Neural Processing Letters, 2022, 54(6): 4971-4986.
[17] KENTON J D M W C, TOUTANOVA L K. BERT: pre-training of deep bidirectional transformers for language understanding[C]//Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers), 2019: 4171-4186.
[18] LIU Y, OTT M, GOYAL N, et al. RoBERTa: a robustly optimized bert pretraining approach[J]. arXiv:1907.11692, 2019.
[19] 黄友文, 魏国庆, 胡燕芳. DistillBIGRU: 基于知识蒸馏的文本分类模型[J]. 中文信息学报, 2022, 36(4): 81-89.
HUANG Y W, WEI G Q, HU Y F. DistillBIGRU: text classification model based on knowledge distillation[J]. Journal of Chinese Information Processing, 2022, 36(4): 81-89.
[20] LIN Y, MENG Y, SUN X, et al. BERTGCN: transductive text classification by combining GCN and BERT[J]. arXiv:2105.05727, 2021.
[21] HU S, DING N, WANG H, et al. Knowledgeable prompt-tuning: Incorporating knowledge into prompt verbalizer for text classification[J]. arXiv:2108.02035, 2021.
[22] 孙坤, 秦博文, 桑基韬, 等. 基于共享语义空间的多标签文本分类[J]. 计算机工程与应用, 2023, 59(12): 100-105.
SUN K, QIN B W, SANG J T, et al. Multi-label text classification based on shared semantic space[J]. Computer Engineering and Applications, 2023, 59(12): 100-105.
[23] 林呈宇, 王雷, 薛聪. 标签语义增强的弱监督文本分类模型[J]. 计算机应用, 2023, 43(2): 335-342.
LIN C Y, WANG L XUE C. Weakly-supervised text classification with label semantic enhancement[J]. Journal of Computer Applications, 2023, 43(2): 335-342.
[24] LIU P, QIU X, HUANG X. Recurrent neural network for text classification with multi-task learning[C]//Proceedings of the International Joint Conference on Artificial Intelligence, New York, USA, 2016: 2873-2879.
[25] VASWANI A, SHAZEER N, PARMAR N, et al. Attention is all you need[C]//Proceedings of the 31st Internatioal Conferenceon Neural Information Processing Systems, 2017: 5998-6008.
[26] 张文轩, 殷雁君, 智敏. 基于概率分布的图卷积文本分类模型[J]. 中文信息学报, 2022, 36(4): 100-110.
ZHANG W X, YIN Y J, ZHI M. Probability distribution based graph convolution network for text classification[J]. Journal of Chinese Information Processing, 2022, 36(4): 100-110.