Abstract
This paper proposes an automatic Chinese text categorization method for solving the emergency event report classification problem. Since the bidirectional encoder representations from transformers (BERT) has achieved great success in the natural language processing domain, it is employed to derive emergency text features in this study. To overcome the data imbalance problem in the distribution of emergency event categories, a novel loss function is proposed to improve the performance of the BERT-based model. Meanwhile, in order to avoid the negative impacts of the extreme learning rate, the Adabound optimization algorithm that achieves a gradual smooth transition from Adam optimizer to stochastic gradient descent optimizer is employed to learn the parameters of the model. The feasibility and competitiveness of the proposed method are validated on both imbalanced and balanced datasets. Furthermore, the generic BERT, BERT ensemble LSTM-BERT (BERT-LB), Attention-based BiLSTM fused CNN with gating mechanism (ABLG-CNN), TextRCNN, Att-BLSTM, and DPCNN are used as benchmarks on these two datasets. Meanwhile, sampling methods, including random sampling, ADASYN, synthetic minority over-sampling techniques (SMOTE), and Borderline-SMOTE, are employed to verify the performance of the proposed loss function on the imbalance dataset. Compared with benchmarking methods, the proposed method has achieved the best performance in terms of accuracy, weighted average precision, weighted average recall, and weighted average F1 values. Therefore, it is promising to employ the proposed method for real applications in smart emergency management systems.
Similar content being viewed by others
References
Li X, Pu W, Zhao X (2019) Agent action diagram: toward a model for emergency management system. Simul Model Pract Theory 94:66–99
De Nicola A, Melchiori M, Villani ML (2019) Creative design of emergency management scenarios driven by semantics: an application to smart cities. Inform Syst 81:21–48
Yu F, Fan B, Li X (2020) Improving emergency preparedness to cascading disasters: A case-driven risk ontology modelling. Journal of Contingencies and Crisis Management p 28
Yao K, Zhang L, Luo T, Wu Y (2018) Deep reinforcement learning for extractive document summarization. Neurocomputing 284(APR.5):52–62
Singh SP et al (2017) Machine translation using deep learning: An overview, pp 162–167
Semberecki P, Maciejewski H (2017) Deep learning methods for subject text classification of articles, pp 357–360
Devlin J, Chang M-W, Lee K, Toutanova K (2018) Bert: Pre-training of deep bidirectional transformers for language understanding. arXiv:1810.04805
Mikolov T, Sutskever I, Chen K, Corrado GS, Dean J (2013) Distributed representations of words and phrases and their compositionality, pp 3111–3119
Mikolov T, Chen K, Corrado G, Dean J (2013) Efficient estimation of word representations in vector space. arXiv:1301.3781
Kim Y (2014) Convolutional neural networks for sentence classification. Eprint Arxiv
Liu P, Qiu X, Huang X (2016) Recurrent neural network for text classification with multi-task learning
Lai S, Xu L, Liu K, Zhao J (2015) Recurrent convolutional neural networks for text classification, AAAI’15, pp 2267–2273 (AAAI Press)
Zhou P, Shi W, Tian J, Qi Z, Xu B (2016) Attention-based bidirectional long short-term memory networks for relation classification, pp 207–212 (Association for Computational Linguistics)
Johnson R, Zhang T (2017) Deep pyramid convolutional neural networks for text categorization, pp 562–570 (Association for Computational Linguistics)
Li Y, Sun G, Zhu Y (2010) Data imbalance problem in text classification, pp 301–305 (IEEE)
Chawla NV, Bowyer KW, Hall LO, Kegelmeyer WP (2002) Smote: synthetic minority over-sampling technique. J Artif Intel Res 16:321–357
Raghuwanshi BS, Shukla S (2020) Smote based class-specific extreme learning machine for imbalanced learning. Knowl Based Syst 187:104814
Liu B, Tsoumakas G (2020) Dealing with class imbalance in classifier chains via random undersampling. Knowl Based Syst 192:105292
Li M, Xiong A, Wang L, Deng S, Ye J (2020) Aco resampling: Enhancing the performance of oversampling methods for class imbalance classification. Knowl Based Syst, p 105818
Cao C, Wang Z (2018) Imcstacking: Cost-sensitive stacking learning with feature inverse mapping for imbalanced problems. Knowl Based Syst 150:27–37
Shi G, Feng C, Xu W, Liao L, Huang H (2020) Penalized multiple distribution selection method for imbalanced data classification. Knowl Based Syst, p 105833
Fernando K, Ruwani M, Tsokos CP (2021) Dynamically weighted balanced loss: class imbalanced learning and confidence calibration of deep neural networks. IEEE Transactions on Neural Networks and Learning Systems
Vaswani A et al (2017) Attention is all you need, pp 5998–6008
Wu Y et al (2016) Google’s neural machine translation system: Bridging the gap between human and machine translation. arXiv:1609.08144
Lin T-Y, Goyal P, Girshick R, He K, Dollár P (2017) Focal loss for dense object detection, pp 2980–2988
Keskar NS, Socher R (2017) Improving generalization performance by switching from adam to sgd. arXiv:1712.07628
Wilson AC, Roelofs R, Stern M, Srebro N, Recht B (2017) The marginal value of adaptive gradient methods in machine learning, pp 4148–4158
Luo L, Xiong Y, Liu Y, Sun X (2019) Adaptive gradient methods with dynamic bound of learning rate (New Orleans, Louisiana)
Liu J, Xia C, Li X, Yan H, Liu T (2020) A bert-based ensemble model for chinese news topic prediction, BDE 2020, pp 18–23 (Association for Computing Machinery, New York, NY USA)
Deng J, Cheng L, Wang Z (2021) Attention-based bilstm fused cnn with gating mechanism model for chinese long text classification. Comput Speech Lang 68:101182
Sun M, Li J, Guo Z, Yu Z, Zheng Y, Si X, Liu Z (2016) Thuctc: an efficient chinese text classifier. GitHub Repository
Moreo A, Esuli A, Sebastiani F (2016) Distributional random oversampling for imbalanced text classification, pp 805–808
Feng H, Dan T, Wang W, Gui R, Liu J, Li Y (2021) A combination of resampling method and machine learning for text classification on imbalanced data, pp 3–17 (springer)
Rupapara V, Rustam F, Shahzad HF, Mehmood A, Ashraf I, Choi GS (2021) Impact of SMOTE on imbalanced text features for toxic comments classification using RVVC model. IEEE Access 9:78621–78634
Han H, Wang W-Y, Mao B-H (2005) Borderline-SMOTE: a new over-sampling method in imbalanced data sets learning, pp 878–887 (Springer)
Acknowledgements
This work was supported in part by the Guangdong Basic and Applied Basic Research Foundation under Grant 2020A1515110431, in part by Scientific and Technological Innovation Foundation of Foshan under Grant BK22BF009, and in part by the National Nature Science and Foundation of China under Grant 62002016.
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Wang, Z., Wang, L., Huang, C. et al. BERT-based chinese text classification for emergency management with a novel loss function. Appl Intell 53, 10417–10428 (2023). https://doi.org/10.1007/s10489-022-03946-x
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10489-022-03946-x