Journal of Computer Applications ›› 2022, Vol. 42 ›› Issue (4): 1125-1130.DOI: 10.11772/j.issn.1001-9081.2021071272
• The 36 CCF National Conference of Computer Applications (CCF NCCA 2020) • Previous Articles Next Articles
Teng TENG, Haiwei PAN(), Kejia ZHANG, Xuelian MU, Ximing ZHANG, Weipeng CHEN
Received:
2021-07-16
Revised:
2022-01-01
Accepted:
2022-01-04
Online:
2022-04-28
Published:
2022-04-10
Contact:
Haiwei PAN
About author:
TENG Teng, born in 1996,M. S. candidate. His research interestsinclude intelligent healthcare,intelligent question-answeringSupported by:
通讯作者:
潘海为
作者简介:
滕腾(1996—),男,黑龙江哈尔滨人,硕士研究生,主要研究方向:智慧医疗、智能问答基金资助:
CLC Number:
Teng TENG, Haiwei PAN, Kejia ZHANG, Xuelian MU, Ximing ZHANG, Weipeng CHEN. Attention mechanism based Stack-CNN model to support Chinese medical questions and answers[J]. Journal of Computer Applications, 2022, 42(4): 1125-1130.
滕腾, 潘海为, 张可佳, 牟雪莲, 张锡明, 陈伟鹏. 支持中文医疗问答的基于注意力机制的栈卷积神经网络模型[J]. 《计算机应用》唯一官方网站, 2022, 42(4): 1125-1130.
Add to citation manager EndNote|Ris|BibTeX
URL: http://www.joca.cn/EN/10.11772/j.issn.1001-9081.2021071272
数据集 | 问题句子数 | 答案句子数 | 每句问题平均字数 | 每句答案平均字数 |
---|---|---|---|---|
总计 | 54 000 | 101 743 | 119 | 212 |
训练集 | 50 000 | 94 134 | 120 | 212 |
开发集 | 2 000 | 3 774 | 117 | 216 |
测试集 | 2 000 | 3 835 | 119 | 211 |
Tab. 1 Statistics of cMedQA dataset
数据集 | 问题句子数 | 答案句子数 | 每句问题平均字数 | 每句答案平均字数 |
---|---|---|---|---|
总计 | 54 000 | 101 743 | 119 | 212 |
训练集 | 50 000 | 94 134 | 120 | 212 |
开发集 | 2 000 | 3 774 | 117 | 216 |
测试集 | 2 000 | 3 835 | 119 | 211 |
编号 | 嵌入方式 | 模型 | 正确率/% | |
---|---|---|---|---|
开发集 | 训练集 | |||
1 | 无 | 随机选择 | 0.10 | 0.10 |
2 | 词匹配(jieba) | 37.05 | 36.60 | |
3 | 词匹配(ICTCLAS) | 35.11 | 36.22 | |
4 | 字匹配 | 33.65 | 34.90 | |
5 | BM25(jieba) | 37.60 | 40.00 | |
6 | BM25(ICTCLAS) | 40.25 | 41.25 | |
7 | BM25(字) | 44.80 | 45.40 | |
8 | 词(jieba) | 平均嵌入 | 15.60 | 16.80 |
9 | 词(ICTCLAS) | 18.05 | 18.75 | |
10 | 字 | 24.90 | 24.00 | |
11 | 词(jieba) | 嵌入匹配 | 24.55 | 23.65 |
12 | 词(ICTCLAS) | 27.85 | 29.10 | |
13 | 字 | 30.80 | 32.30 |
Tab. 2 Accuracy comparison of traditional baseline models
编号 | 嵌入方式 | 模型 | 正确率/% | |
---|---|---|---|---|
开发集 | 训练集 | |||
1 | 无 | 随机选择 | 0.10 | 0.10 |
2 | 词匹配(jieba) | 37.05 | 36.60 | |
3 | 词匹配(ICTCLAS) | 35.11 | 36.22 | |
4 | 字匹配 | 33.65 | 34.90 | |
5 | BM25(jieba) | 37.60 | 40.00 | |
6 | BM25(ICTCLAS) | 40.25 | 41.25 | |
7 | BM25(字) | 44.80 | 45.40 | |
8 | 词(jieba) | 平均嵌入 | 15.60 | 16.80 |
9 | 词(ICTCLAS) | 18.05 | 18.75 | |
10 | 字 | 24.90 | 24.00 | |
11 | 词(jieba) | 嵌入匹配 | 24.55 | 23.65 |
12 | 词(ICTCLAS) | 27.85 | 29.10 | |
13 | 字 | 30.80 | 32.30 |
模型 | 正确率 | |
---|---|---|
开发集 | 测试集 | |
Multi-CNNs | 48.40 | 51.15 |
Stack-CNN | 46.03 | 47.62 |
Att-StackCNN | 46.22 | 47.60 |
Tab. 3 Accuracy comparison of deep learning methods based on word (jieba)
模型 | 正确率 | |
---|---|---|
开发集 | 测试集 | |
Multi-CNNs | 48.40 | 51.15 |
Stack-CNN | 46.03 | 47.62 |
Att-StackCNN | 46.22 | 47.60 |
模型 | 正确率 | |
---|---|---|
开发集 | 测试集 | |
Multi-CNNs | 53.06 | 52.34 |
Stack-CNN | 53.07 | 52.24 |
Att-StackCNN | 53.12 | 52.31 |
Tab. 4 Accuracy comparison of deep learning methods based on word (ICTCLAS)
模型 | 正确率 | |
---|---|---|
开发集 | 测试集 | |
Multi-CNNs | 53.06 | 52.34 |
Stack-CNN | 53.07 | 52.24 |
Att-StackCNN | 53.12 | 52.31 |
模型 | 正确率 | |
---|---|---|
开发集 | 测试集 | |
Multi-CNNs | 64.20 | 65.85 |
Stack-CNN | 63.50 | 65.20 |
Att-StackCNN | 64.83 | 66.35 |
Tab. 5 Comparison of correct rate of deep learning methods based on character embedding
模型 | 正确率 | |
---|---|---|
开发集 | 测试集 | |
Multi-CNNs | 64.20 | 65.85 |
Stack-CNN | 63.50 | 65.20 |
Att-StackCNN | 64.83 | 66.35 |
卷积核 尺度 | 正确率 | 卷积核 尺度 | 正确率 | ||
---|---|---|---|---|---|
开发集 | 测试集 | 开发集 | 测试集 | ||
(2,3) | 63.45 | 64.68 | (2,4) | 64.55 | 65.54 |
(3,4) | 64.57 | 65.52 | (2,3,4) | 64.83 | 66.35 |
Tab. 6 Accuracy comparison of different convolution kernels in Att-StackCNN
卷积核 尺度 | 正确率 | 卷积核 尺度 | 正确率 | ||
---|---|---|---|---|---|
开发集 | 测试集 | 开发集 | 测试集 | ||
(2,3) | 63.45 | 64.68 | (2,4) | 64.55 | 65.54 |
(3,4) | 64.57 | 65.52 | (2,3,4) | 64.83 | 66.35 |
1 | 于倩倩. 面向医疗领域的中文自动问答系统的设计与实现[D]. 北京:北京邮电大学, 2020: 1-3. |
YU Q Q. Design and implementation of Chinese automatic question answering system oriented medical field[D]. Beijing: Beijing University of Posts and Telecommunications, 2020:1-3. | |
2 | Class_guy. 问答系统综述[EB/OL]. (2018-08-09) [2021-07-14].[EB/OL]. (2018-08-09) [2021-07-14].). 10.1002/9781119209164 |
3 | HSU W N, ZHANG Y, GLASS J. Recurrent neural network encoder with attention for community question answering[EB/OL]. (2016-03-23) [2021-07-14].. 10.18653/v1/s16-1128 |
4 | WANG J, MAN C T, ZHAO Y F, et al. An answer recommendation algorithm for medical community question answering systems[C]// Proceedings of the 2016 IEEE International Conference on Service Operations and Logistics, and Informatics. Piscataway: IEEE, 2016: 139-144. 10.1109/soli.2016.7551676 |
5 | LI T C, HAO Y, ZHU X Y, et al. A Chinese question answering system for specific domain[C]// Proceedings of the 2014 International Conference on Web-Age Information Management, LNCS 8485. Cham: Springer, 2014: 590-601. |
6 | WANG B Y, NIU J B, MA L Q, et al. A Chinese question answering approach integrating count-based and embedding-based features[C]// Proceedings of the 2016 International Conference on Computer Processing of Oriental Languages/ National CCF Conference on Natural Language Processing and Chinese Computing. Cham: Springer, 2016: 934-941. 10.1007/978-3-319-50496-4_88 |
7 | ZHANG S, ZHANG X, WANG H, et al. Chinese medical question answer matching using end-to-end character-level multi-scale CNNs[J]. Applied Sciences, 2017, 7(8): No.767. 10.3390/app7080767 |
8 | ZHANG Y T, LU W P, OU W H, et al. Chinese medical question answer matching with stack-CNN[C]// Proceedings of the 2018 International Symposium on Artificial Intelligence and Robotics, SCI 810. Cham: Springer, 2018: 455-462. |
9 | BENGIO Y, DUCHARME R, VINCENT P, et al. A neural probabilistic language model[J]. Journal of Machine Learning Research, 2003, 3: 1137-1155. 10.1007/3-540-33486-6_6 |
10 | MIKOLOV T, SUTSKEVER I, CHEN K, et al. Distributed representations of words and phrases and their compositionality[C]// Proceedings of the 26th International Conference on Neural Information Processing Systems. Red Hook, NY: Curran Associates Inc., 2013: 3111-3119. |
11 | TADDY M. Document classification by inversion of distributed language representations[C]// Proceedings of the 53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Conference on Natural Language Processing (Volume 2: Short Papers). Stroudsburg, PA: Association for Computational Linguistics, 2015: 45-49. 10.3115/v1/p15-2008 |
12 | HUANG C C, QIU X P, HUANG X J. Text classification with document embeddings[C]// Proceedings of the 2014 International Symposium on Natural Language Processing Based on Naturally Annotated Big Data/ China National Conference on Chinese Computational Linguistics, LNCS 8801. Cham: Springer, 2014: 131-140. |
13 | LEVY O, GOLDBERG Y. Neural word embedding as implicit matrix factorization[C]// Proceedings of the 27th International Conference on Neural Information Processing Systems. Cambridge: MIT Press, 2014: 2177-2185. |
14 | ZHANG X, ZHAO J B, LeCUN Y. Character-level convolutional networks for text classification[C]// Proceedings of the 28th International Conference on Neural Information Processing Systems. Cambridge: MIT Press, 2015: 647-657. 10.1109/icip.2015.7351229 |
15 | KIM Y. Convolutional neural networks for sentence classification[C]// Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing. Stroudsburg, PA: Association for Computational Linguistics, 2014: 1746-1751. 10.3115/v1/d14-1181 |
16 | YIN W P, SCHÜTZE H, XIANG B, et al. ABCNN: attention-based convolutional neural network for modeling sentence pairs[J]. Transactions of the Association for Computational Linguistics, 2016, 4: 259-272. 10.1162/tacl_a_00097 |
17 | GOEURIOT L, JONES G J F, KELLY L, et al. Medical information retrieval: introduction to the special issue[J]. Information Retrieval Journal, 2016, 19(1/2): 1-5. 10.1007/s10791-015-9277-8 |
18 | MANNING C D, PRABHAKAR R, SCHÜTZE H. Introduction to Information Retrieval[M]. Cambridge: Cambridge University Press, 2008:151-177. 10.1017/cbo9780511809071 |
19 | 李岩,郭凤英,翟兴,等. 基于jieba中文分词的在线医疗网站医生画像研究[J]. 医学信息学杂志, 2020, 41(7): 14-18. 10.3969/j.issn.1673-6036.2020.07.003 |
LI Y, GUO F Y, ZHAI X, et al. Study on doctors’ portraits of online medical website based on jieba Chinese word segmentation[J]. Journal of Medical Informatics, 2020, 41(7): 14-18. 10.3969/j.issn.1673-6036.2020.07.003 | |
20 | 徐玮. 医疗问答系统的中文分词算法研究[D]. 武汉:华中科技大学, 2019: 7-8. |
XU W. The study of the Chinese word segmentation algorithm in medical question answering system[D]. Wuhan: Huazhong University of Science and Technology, 2019:7-8. | |
21 | CUI Y M, LIU T, CHEN Z P, et al. Consensus attention-based neural networks for Chinese reading comprehension[C]// Proceedings of the 26th International Conference on Computational Linguistics: Technical Papers. Stroudsburg, PA: Association for Computational Linguistics, 2016: 1777-1786. 10.18653/v1/p17-1055 |
22 | YU L, HERMANN K M, BLUNSOM P, et al. Deep learning for answer sentence selection[EB/OL]. (2014-12-04) [2021-07-14].. |
[1] | Jin ZHANG, Peiqi QU, Cheng SUN, Meng LUO. Safety helmet wearing detection algorithm based on improved YOLOv5 [J]. Journal of Computer Applications, 2022, 42(4): 1292-1300. |
[2] | Shoulong JIAO, Youxiang DUAN, Qifeng SUN, Zihao ZHUANG, Chenhao SUN. Knowledge representation learning method incorporating entity description information and neighbor node features [J]. Journal of Computer Applications, 2022, 42(4): 1050-1056. |
[3] | Yongfeng DONG, Yuehua SUN, Lichao GAO, Peng HAN, Haipeng JI. Fault diagnosis method based on improved one-dimensional convolutional and bidirectional long short-term memory neural networks [J]. Journal of Computer Applications, 2022, 42(4): 1207-1215. |
[4] | Junhua GU, Shuai FAN, Ningning LI, Suqi ZHANG. Long- and short-term recommendation model and updating method based on knowledge graph preference attention network [J]. Journal of Computer Applications, 2022, 42(4): 1079-1086. |
[5] | Zhihua LIU, Wenjie CHEN, Aibin CHEN. Homologous spectrogram feature fusion with self-attention mechanism for bird sound classification [J]. Journal of Computer Applications, 2022, 42(4): 1260-1268. |
[6] | Junhua GU, Rui WANG, Ningning LI, Suqi ZHANG. Knowledge graph attention network fusing collaborative filtering information [J]. Journal of Computer Applications, 2022, 42(4): 1087-1092. |
[7] | Wangjing TANG, Bin XU, Meihan TONG, Meihuan HAN, Liming WANG, Qi ZHONG. Popular science text classification model enhanced by knowledge graph [J]. Journal of Computer Applications, 2022, 42(4): 1072-1078. |
[8] | Xinrong HU, Junyu ZHANG, Tao PENG, Junping LIU, Ruhan HE, Kai HE. Cascaded cross-domain feature fusion for virtual try-on [J]. Journal of Computer Applications, 2022, 42(4): 1269-1274. |
[9] | Wenjing JIANG, Xi XIONG, Zhongzhi LI, Binyong LI. Recommendation system based on non-sampling collaborative knowledge graph network [J]. Journal of Computer Applications, 2022, 42(4): 1057-1064. |
[10] | Tingxiu CHEN, Jianqin YIN. Audio visual joint action recognition based on key frame selection network [J]. Journal of Computer Applications, 2022, 42(3): 731-735. |
[11] | Wenqiu ZHU, Guang ZOU, Zhigao ZENG. Object tracking algorithm with hierarchical features and hybrid attention [J]. Journal of Computer Applications, 2022, 42(3): 833-843. |
[12] | Na YU, Yan LIU, Xiongju WEI, Yuan WAN. Semantic segmentation of RGB-D indoor scenes based on attention mechanism and pyramid fusion [J]. Journal of Computer Applications, 2022, 42(3): 844-853. |
[13] | Mao WANG, Yaxiong PENG, Anjiang LU. Cross-modal chiastopic-fusion attention network for visual question answering [J]. Journal of Computer Applications, 2022, 42(3): 854-859. |
[14] | Yuming ZHAO, Shenkai GU. Adversarial attack defense model with residual dense block self-attention mechanism and generative adversarial network [J]. Journal of Computer Applications, 2022, 42(3): 921-929. |
[15] | Pengwei LIU, Yuan GAO, Pinle QIN, Zhe YIN, Lifang WANG. Medical MRI image super-resolution reconstruction based on multi-receptive field generative adversarial network [J]. Journal of Computer Applications, 2022, 42(3): 938-945. |
Viewed | ||||||
Full text |
|
|||||
Abstract |
|
|||||