Clinical Relation Extraction via Dual Piecewise Attention Neural Tensor Network

被引：0

作者：

Wei H. ^{[1
,2
]}

Tang H.-L. ^{[3
]}

Zhou A. ^{[2
]}

Zhang Y.-J. ^{[2
]}

Chen F. ^{[2
]}

Lu M.-Y. ^{[2
]}

机构：

[1] School of Software, Dalian University of Foreign Languages, Liaoning, Dalian

[2] Information Science and Technology College, Dalian Maritime University, Liaoning, Dalian

[3] School of Computer Science and Technology, Shandong Technology and Business University, Shandong, Yantai

来源：

Tien Tzu Hsueh Pao/Acta Electronica Sinica | 2023年 / 51卷 / 03期

基金：

中国国家自然科学基金;

关键词：

clinical texts; neural tensor network; piecewise attention mechanism; relation extraction; sample imbal⁃ ance;

D O I：

10.12263/DZXB.20210628

中图分类号：

学科分类号：

摘要：

At present, biomedical relation extraction has made considerable progress. However, when dealt with com⁃ plex clinical texts, due to the large number of long sentences and the high density distribution of entity pairs in the sentenc⁃ es, the existing methods of relation extraction still have defects. We propose a relation extraction model via tensor-based bi⁃ directional gate recurrent unit (Tensor-BiGRU) and piecewise attention mechanism. The ability of BiGRU to extract the un⁃ derlying features is enhanced based on tensor weight matrix. Two kinds of piecewise attention mechanisms are proposed to improve the performance of the model in capturing long sentence features.When the sentence has multiple entity pairs, the semantic representations of the entity pairs are introduced to overcome the performance degradation of the mode. A weight-adaptive cross-entropy loss function is proposed to improve the generalization of the model when the sample distribution of different relation categories in the dataset is unbalanced. The experimental results show that without relying on any feature engineering and high-performance computing environment, the model achieves advanced performance on the 2010 i2b2/VA clinical data set. © 2023 Chinese Institute of Electronics. All rights reserved.

引用

页码：658 / 665

页数：7

共 16 条

[1] LI Z X, SUN Y R, TANG S Q, Et al., Dual attention guided graph convolutional networks for relation extraction, Ac⁃ ta Electronica Sinica, 49, 2, pp. 315-323, (2021)
[2] FENG J Z, SONG S S, WANG Y Z, Et al., Entity relation extraction based on improved attention mechanism, Ac⁃ ta Electronica Sinica, 47, 8, pp. 1692-1700, (2019)
[3] YANG J F, YU Q B, GUAN Y, Et al., An overview of re⁃ search on electronic medical record oriented named entity recognition and entity relation extraction, Acta Automat⁃ ica Sinica, 40, 8, pp. 1537-1562, (2014)
[4] RINK B, HARABAGIU S, ROBERTS K., Automatic ex⁃ traction of relations between medical concepts in clinical texts, Journal of the American Medical Informatics As⁃ sociation, 18, 5, pp. 594-600, (2011)
[5] DE BRUIJN B, CHERRY C, KIRITCHENKO S, Et al., Ma⁃ chine-learned solutions for three stages of clinical informa⁃ tion extraction: The state of the art at i2b2 2010[J], Journal of the American Medical Informatics Association, 18, 5, pp. 557-562, (2011)
[6] SAHU S K, ANAND A, ORUGANTY K, Et al., Relation extraction from clinical texts using domain invariant con⁃ volutional neural network
[7] LUO Y., Recurrent neural networks for classifying rela⁃ tions in clinical notes, Journal of Biomedical Informat⁃ ics, 72, pp. 85-95, (2017)
[8] HE B, GUAN Y, DAI R., Classifying medical relations in clinical text via convolutional neural networks, Artifi⁃ cial Intelligence in Medicine, 93, pp. 43-49, (2019)
[9] LI Z, YANG J S, GOU X, Et al., Recurrent neural networks with segment attention and entity description for relation extraction from clinical texts, Artificial Intelligence in Medicine, 97, pp. 9-18, (2019)
[10] NING S M, TENG F, LI T R., Multi-channel self-atten⁃ tion mechanism for relation extraction in clinical records, Chinese Journal of Computers, 43, 5, pp. 916-929, (2020)

← 1 2 →