School of Computer Science and Engineering,Northeastern University,Shenyang 110169,China. Corresponding author: ZHU Hong-zhao,E-mail: 1024921337@qq. com
To enhance the comprehensiveness and accuracy of the medical decision support knowledge graph, a collective entity disambiguation method based on structured gradient tree boosting (CED-SGTB) was proposed to address challenges in the medical domain, including long, complex, and highly specialized entity texts, as well as the ambiguity of the same entity across different databases. Firstly, medical decision texts were cleaned and processed, and neural networks were deployed to collaboratively detect and annotate entities within the texts. Secondly, global entity disambiguation was achieved by optimizing the current candidate entities and previous entities through global features. Thirdly, a golden path bidirectional beam search algorithm was designed to reduce model variance. Finally, the results of comparative experiments and ablation experiments demonstrate that CED-SGTB outperforms traditional methods in terms of accuracy and F1-score and can more accurately complete entity disambiguation tasks.
KörschensM, BucherS F, BodesheimP, et al. Determining the community composition of herbaceous species from images using convolutional neural networks[J]. Ecological Informatics, 2024, 80: 102516.
[2]
QianY, PanL. Leveraging multimodal features for knowledge graph entity alignment based on dynamic self-attention networks[J]. Expert Systems with Applications, 2023, 228: 120363.
[3]
RanY Y, XuX B, LuoM Z, et al. Scene classification method based on multi-scale convolutional neural network with long short-term memory and whale optimization algorithm[J]. Remote Sensing, 2024, 16(1): 174.
[4]
LiuX, ZhangL, ZhengQ S, et al. Construction of an event knowledge graph based on a dynamic resource scheduling optimization algorithm and semantic graph convolutional neural networks[J]. Electronics, 2024, 13(1): 11.
[5]
NafaY, ChenQ, HouB Y, et al. Adaptive deep learning for entity disambiguation via knowledge-based risk analysis[J]. Expert Systems with Applications, 2024, 238: 122342.
[6]
BasileA, CrupiR, GrassoM, et al. Disambiguation of company names via deep recurrent networks[J]. Expert Systems with Applications, 2024, 238: 122035.
[7]
WuW M, HuJ T, ZhangZ J, et al. Deterministic learning-based neural identification and knowledge fusion[J]. Neural Networks, 2023, 166: 688-702.
[8]
WeiF F, MeiK Z. Frequency inception based graph neural network for relation prediction in knowledge graphs[J]. Knowledge-Based Systems, 2023, 278: 110908.
ZhangJi-xiang, ZhangXiang-sen, WuChang-xu, et al. Review of knowledge graph construction techniques[J]. Computer Engineering, 2022, 48(3): 23-37.
[18]
WangC X, HuangZ H, WanY, et al. FuAlign: cross-lingual entity alignment via multi-view representation learning of fused knowledge graphs[J]. Information Fusion, 2023, 89: 41-52.
[19]
BouarroudjW, BoufaidaZ, BellatrecheL. Named entity disambiguation in short texts over knowledge graphs[J]. Knowledge and Information Systems, 2022, 64(2): 325-351.
[20]
ZhangZ R, YangY Y, ChenB H. Relation-aware heterogeneous graph neural network for entity alignment[J]. Neurocomputing, 2024, 592: 127797.
[21]
YangS Y, ZhangP L, CheC, et al. B-LBConA: a medical entity disambiguation model based on Bio-LinkBERT and context-aware mechanism[J]. BMC Bioinformatics, 2023, 24(1): 97.
[22]
ZhuB B, BaoT, WangK R, et al. A semi-supervised neighborhood matching model for global entity alignment[J]. Neural Computing and Applications, 2023, 35(15): 10779-10799.
WenPing-mei, YeZhi-wei, DingWen-jian, et al. A review of research progress on named entity disambiguation[J]. Data Analysis and Knowledge Discovery, 2020, 4(9): 15-25.
[25]
LengJ, TangX J. Graph attention networks for multiple pairs of entities and aspects sentiment analysis in long texts[J]. Journal of Systems Science and Information, 2022, 10(3): 203-215.
[26]
ZhuL, LiN, BaiL Y. Embedding-based entity alignment between multi-source temporal knowledge graphs[J]. Engineering Applications of Artificial Intelligence, 2024, 133: 108451.
[27]
HuP H, YeQ, ZhangW Y, et al. Integration of multiple terminology bases: a multi-view alignment method using the hierarchical structure[J]. Bioinformatics, 2023, 39(11): 689.
[28]
MaX J, LiZ Y, SongG J, et al. Learning discrete adaptive receptive fields for graph convolutional networks[J]. Science China Information Sciences, 2023, 66(12): 222101.
[29]
LinL, ZuL Z, GuoF, et al. Using combinatorial optimization to solve entity alignment: an efficient unsupervised model[J]. Neurocomputing, 2023, 558: 126802.
DuanZong-tao, LiFei, ChenZhe. A survey of entity disambiguation[J]. Control and Decision, 2021, 36(5): 1025-1039.
[32]
ZhangY X, LiY X, WeiX, et al. Graph matching for knowledge graph alignment using edge-coloring propagation[J]. Pattern Recognition, 2023, 144: 109851.
[33]
XieC X, ZhangL M, ZhongZ G. Entity alignment method based on joint learning of entity and attribute representations[J]. Applied Sciences, 2023, 13(9): 5748.
[34]
WangM, ShiY H, YangH, et al. Probing the impacts of visual context in multimodal entity alignment[J]. Data Science and Engineering, 2023, 8(2): 124-134.