In order to address problems related to label dependency, imbalance, extensive text analysis, and multi-label classification in legal contexts, this paper presents the BERT-CNN framework. The model’s development and integration with actual court cases allowed for the identification of the ideal hyperparameter configuration for the BERT-CNN architecture. The gathered dataset can be divided into three separate sections for the purpose of conducting a thorough assessment and examination of the model: the training, validation, and testing sets. The evaluation data from the testing set is used to evaluate the model’s performance. The evaluation’s conclusions show that the BERT-CNN model obtains metrics for precision, recall, F1 score and AUC-PR of 0.83, 0.82, 0.83 and 0.88, in that order. The empirical results highlight the BERT-CNN framework’s applicability for classifying legal texts, demonstrating impressive resilience and precision in multi-label classification tasks pertaining to court cases.
KeyueLYiCNiuS. Text classification algorithm for social e-commerce based on BERT. Comp Sci2021; 048(002): 87–92.
2.
ÇepniSToprakAGYatkınogluA, et al.Performance evaluation of a pretrained BERT Model for automatic text classification. J Artifi Intell Data Sci2023; 3(1): 27–35.
3.
LiuNWangQRenJ. Label-embedding bi-directional attentive model for multi-label text classification. Neural Process Lett2021; 53: 375–389.
4.
SunCQiuXXuY, et al.How to fine-tune bert for text classification?In: China national conference on Chinese computational linguistics. Cham: Springer International Publishing, October, 2019, pp. 194–206.
5.
TangXSHaoKWeiH. A bio-inspired positional embedding network for transformer-based models. Neural Networks2023;166: 204-214.
6.
PingYWangB. Retranslated Chinese classical canon Journey to the West: a stylometric comparison between Julia Lovell’s retranslation and Arthur Waley’s translation. Digital Scholarship in the Humanities2024; 39(1): 308–320.
7.
ChenXCongPLvS. A long-text classification method of Chinese news based on BERT and CNN. IEEE Access2022; 10: 34046–34057.
8.
BaoTRenNLuoR, et al.A BERT-based hybrid short text classification model incorporating CNN and attention-based BiGRU. J Organ End User Comput2021; 33(6): 1–21.
9.
CaiLSongYLiuT, et al.A hybrid BERT model that incorporates label semantics via adjustive attention for multi-label text classification. IEEE Access2020; 8: 152183–152192.
10.
IslamMT. Plant disease detection using CNN model and image processing. Int J Eng Res Technol2020; 9(10): 291–297.
11.
ChungYLChungHYTsaiWF. Hand gesture recognition via image processing techniques and deep CNN. J Intell Fuzzy Syst2020; 39(3): 4405–4418.
12.
TripathiM. Analysis of convolutional neural network based image classification techniques. J Innov Imag Proces (JIIP)2021; 3(02): 100–117.
13.
WangHHeJZhangX, et al.A short text classification method based on N‐gram and CNN. Chin J Electron2020; 29(2): 248–254.
14.
DharmaEMGaolFLWarnarsH, et al.The accuracy comparison among word2vec, glove, and fasttext towards convolution neural network (cnn) text classification. J Theor Appl Inf Technol2022; 100(2): 349–359.
15.
ZhaoWZhuLWangM, et al.WTL-CNN: a news text classification method of convolutional neural network based on weighted word embedding. Connect Sci2022; 34(1): 2291–2312.
16.
SharmaAKChaurasiaSSrivastavaDK. Sentimental short sentences classification by using CNN deep learning model with fine tuned Word2Vec. Procedia Comput Sci2020; 167: 1139–1147.
17.
LiangYLiHGuoB, et al.Fusion of heterogeneous attention mechanisms in multi-view convolutional neural network for text classification. Inf Sci2021; 548: 295–312.
18.
MinaeeSKalchbrennerNCambriaE, et al.Deep learning--based text classification: a comprehensive review. ACM Comput Surv2021; 54(3): 1–40.
19.
ChenXZouM. Research on short text classification based on the BERT-BiGRU-CNN model. J Jilin Univ (Inform Sci Ed)2023; 41(6): 1048–1053.
20.
YeRShaoJZhangX. Research on the knowledge distillation method of news text classification based on BERT-CNN. Electron Tech App2023; 49(1): 8–13.
21.
SaadahSAuditamaKMFattahilaAA, et al.Implementation of BERT, IndoBERT, and CNN-LSTM in classifying public opinion about COVID-19 vaccine in Indonesia. Jurnal RESTI (Rekayasa Sistem dan Teknologi Informasi)2022; 6(4): 648–655.
22.
ZhangMZhouHCaoJ. Research on the classification of double-BERT directed emotional texts based on attention mechanism. J Intell Syst2022; 17(6): 1220–1227.
23.
SelvaBSKanniga DeviR. A review on word embedding techniques for text classification. Innov Data Commun Tech App: Proce of ICIDCA2020; 2021: 267–281.
24.
BilalMAlmazroiAA. Effectiveness of fine-tuned BERT model in classification of helpful and unhelpful online customer reviews. Electron Commer Res2023; 23(4): 2737–2757.
25.
WanCXLiB. Financial causal sentence recognition based on BERT-CNN text classification. J Supercomput2022; 78: 1–25.
26.
JacobIJ. Performance evaluation of caps-net based multitask learning architecture for text classification. J Artif Intell2020; 2(01): 1–10.
27.
ManoharanJS. Capsule network algorithm for performance optimization of text classification. J Soft Comput Para (JSCP)2021; 3(01): 1–9.
28.
SoniSChouhanSSRathoreSS. TextConvoNet: a convolutional neural network based architecture for text classification. Appl Intell2023; 53(11): 14249–14268.
29.
ZulqarnainMGhazaliRHassimYMM, et al.Text classification based on gated recurrent unit combines with support vector machine. Int J Electr Comput Eng2020; 10(4): 3734–3742.
30.
KhasanahIN. Sentiment classification using fasttext embedding and deep learning model. Procedia Comput Sci2021; 189: 343–350.
