Literature mirrors a nation’s ideology and language system, with British and American literature (BALW) differing greatly from Chinese literature due to distinct historical, linguistic, and cultural backgrounds. Chinese language system struggles to accurately retrieve and classify BALW. To tackle this, we propose a BALW appreciation model. Built upon a base model akin to DocBERT, it uses the BERT pre-trained model in the text representation module to capture contextual semantics. A heterogeneous graph attention network (HAN) with word, feature word, and label nodes is designed to extract local semantics in the text. These features are then integrated for multi-label classification. Experiments on a curated dataset show our model outperforms the base one, improving Hamming Loss, Macro-F1, and Micro-F1 by 0.009, 1.9%, and 5.38%, respectively. This enhances intelligent classification and retrieval of BALW, benefiting literary appreciation and cross-cultural literary exchange.
YangYZhangJLiuD, et al.Low-rank and sparse matrix decomposition with background position estimation for hyperspectral anomaly detection. Infrared Phys Technol2018; 96: 213–227.
12.
SunLYinTDingW, et al.Feature selection with missing labels using multilabel fuzzy neighborhood rough sets and maximum relevance minimum redundancy. IEEE Trans Fuzzy Syst2021; 30(5): 1197–1211.
13.
El-AbsiMZhengFAbuelhaijaA, et al.Indoor large-scale MIMO-based RSSI localization with low-complexity RFID infrastructure. Sensors2020; 20(14): 3933.
14.
PengYYanSLuZ. Transfer learning in biomedical natural language processing: an evaluation of BERT and ELMo on ten benchmarking datasets, 2019. arXiv preprint arXiv:1906.05474.
15.
MaS. A study of two-way short- and long-term memory network intelligent computing IoT model-assisted home education attention mechanism. Comput Intell Neurosci2021; 2021: 2021.
16.
GautamAVenkteshVMasudS. Fake news detection system using xlnet model with topic distributions: constraint@ aaai2021 shared task. In: Combating Online Hostile Posts in Regional Languages during Emergency Situation: First International Workshop, CONSTRAINT 2021, Collocated with AAAI 2021. Virtual Event, February 8, 2021 Revised selected papers 1. New York: Springer International Publishing, 2021, pp. 189–200.
17.
MengFYangSWangJ, et al.Creating knowledge graph of electric power equipment faults based on BERT-BiLSTM-CRF model. J Electr Eng Technol2022; 17(4): 2507–2516.
18.
LiaoWWangYYinY, et al.Improved sequence generation model for multi-label classification via CNN and initialized fully connection. Neurocomputing2020; 382: 188–195.
19.
KattenbornTLeitloffJSchieferF, et al.Review on convolutional neural networks (CNN) in vegetation remote sensing. ISPRS J Photogrammetry Remote Sens2021; 173: 24–49.
20.
CaoJYangGYangX. A pixel-level segmentation convolutional neural network based on deep feature fusion for surface defect detection. IEEE Trans Instrum Meas2020; 70: 1–12.
21.
KongDWuF. HST-LSTM: a hierarchical spatial-temporal long-short term memory network for location prediction[C]. IJCAI2018; 18(7): 2341–2347.
22.
NiuZZhongGYuH. A review on the attention mechanism of deep learning. Neurocomputing2021; 452: 48–62.
23.
SoydanerD. Attention mechanism in neural networks: where it comes and where it goes. Neural Comput Appl2022; 34(16): 13371–13385.
24.
VeličkovićPCucurullGCasanovaA, et al.Graph attention networks, 2017. arXiv preprint arXiv:1710.10903.
25.
JinXXiXZhouD, et al.An unsupervised multi-focus image fusion method based on Transformer and U-Net. IET Image Process2023; 17(3): 733–746.
26.
HuangQYuJWuJ, et al.Heterogeneous graph attention networks for early detection of rumors on twitter. In: 2020 international joint conference on neural networks (IJCNN). Piscataway: IEEE, 2020, pp. 1–8.
27.
ZhangXGweonHProvostS. Threshold moving approaches for addressing the class imbalance problem and their application to multi-label classification. In: 2020 4th international conference on advances in image processing. New York: Association for Computing Machinery, 2020, pp. 72–77.
