Hybrid Neural Network for Recognition of the “de” Structure with Semantic Ellipsis

doi:10.13209/j.0479-8023.2018.058

Acta Scientiarum Naturalium Universitatis Pekinensis ›› 2019, Vol. 55 ›› Issue (1): 75-83.DOI: 10.13209/j.0479-8023.2018.058

Previous Articles Next Articles

Hybrid Neural Network for Recognition of the “de” Structure with Semantic Ellipsis

SHI Bingqing¹, DAI Rubing², QU Weiguang^1,2,3,†, GU Yanhui¹, ZHOU Junsheng¹, LI Bin², XU Ge³, SHI Shengwang¹

1. School of Computer Science and Technology, Nanjing Normal University, Nanjing 210023
2. School of Chinese Language and Literature, Nanjing Normal University, Nanjing 210097
3. Fujian Provincial Key Laboratory of Information Processiong and Intelligent Control, Minjiang University, Fuzhou 350121

Received:2018-04-15 Revised:2018-08-15 Online:2019-01-20 Published:2019-01-20
Contact: QU Weiguang, E-mail: wgqu_nj(at)163.com

基于组合神经网络的语义省略“的”字结构识别

侍冰清¹, 戴茹冰², 曲维光^1,2,3,†, 顾彦慧¹, 周俊生¹, 李斌², 徐戈³, 史胜旺¹

1. 南京师范大学计算机科学与技术学院, 南京 210023
2. 南京师范大学文学院, 南京 210097
3. 闽江学院福建省信息处理与智能控制重点实验室, 福州 350121

通讯作者: 曲维光, E-mail: wgqu_nj(at)163.com
基金资助:
国家自然科学基金(61772278, 61472191)、江苏省高校哲学社会科学优秀创新团队项目(2017STD006)和福建省信息处理与智能控制重点实验室开放基金(MJUKF201705)资助

Abstract

Abstract:

To slove the classification of the “de” structure containing the usage of semantic ellipsis, a hybrid neural network is built. Firstly, the network uses a bidirectional LSTM (long short-term memory) neural network to learn more syntactic and semantic information of the “de” structure. Then, the network employs a Max-pooling
layer or GRU (gated recurrent unit) based multiple attention layers to capture features of ellipsis of the “de” structure by which the network can recognize the “de” structure containing the usage of semantic ellipsis. Experiments on CTB8.0 corpus show that the proposed approach can achieve accurate results efficiently, the F1 value is 96.67%.

Key words: neural network, “de&rdquo, structure, semantic ellipsis

摘要：

针对语义省略“的”字结构识别任务, 提出一种基于组合神经网络的识别方法。利用词语和词性, 通过双向LSTM (long short-term memory)神经网络, 学习“的”字结构深层次的语义语法表示。通过Max-pooling层和基于GRU(gated recurrent unit)的多注意力层, 捕获“的”字结构的省略特征, 完成语义省略“的”字结构识别任务。实验结果表明, 所提模型在CTB8.0(Chinese Treebank 8.0)语料中, 能够有效地识别语义省略的“的”字结构, F1值达到96.67%。

关键词: 神经网络, “的”字结构, 语义省略

SHI Bingqing, DAI Rubing, QU Weiguang, GU Yanhui, ZHOU Junsheng, LI Bin, XU Ge, SHI Shengwang. Hybrid Neural Network for Recognition of the “de” Structure with Semantic Ellipsis[J]. Acta Scientiarum Naturalium Universitatis Pekinensis, 2019, 55(1): 75-83.

侍冰清, 戴茹冰, 曲维光, 顾彦慧, 周俊生, 李斌, 徐戈, 史胜旺. 基于组合神经网络的语义省略“的”字结构识别[J]. 北京大学学报自然科学版, 2019, 55(1): 75-83.

[1]	PENG Mou, LI Jianghai, YANG Bo. Pore Structure Characteristics and Influencing Factors of Deep Sandstone Reservoirs: A Case Study of Guaizihu Depression in Yin’e Basin [J]. Acta Scientiarum Naturalium Universitatis Pekinensis, 2024, 60(2): 249-264.
[2]	WANG Bingxue, WU Zongzhi, CHEN Tianyi, LIU Shufeng, LIANG Enhang, LIU Tang. Characterization of the Phage Community Structure in Typical Tributaries of the Yellow River [J]. Acta Scientiarum Naturalium Universitatis Pekinensis, 2024, 60(2): 306-314.
[3]	LI Dai, WANG Tianmu, ZHANG Si, QIN Yue, XIE Fugui, LIU Xinjun, NIE Zhenguo, HUANG Hongshi. Intelligent Diagnosis on Anterior Cruciate Ligament Deficiency Based on Plantar Pressure and ConvLSTM Neural Network [J]. Acta Scientiarum Naturalium Universitatis Pekinensis, 2024, 60(1): 109-117.
[4]	ZHANG Kuo, WANG Lina, FENG Yiming, ZHOU Jianhang, LI Xinjue, LIU Sitong. Effects of Refrigerate Incubation on Nitrogen Removal Performance and Microbial Community of Anammox [J]. Acta Scientiarum Naturalium Universitatis Pekinensis, 2023, 59(6): 1003-1010.
[5]	LIU Jiaju, LI Jincheng, GUO Huaicheng, YUAN Peng, LI Zheng, ZHANG Yang, WANG Zhiyong. Study on Hydrochemical Change Trend of Yarlung Tsangpo River Based on Artificial Neural Network [J]. Acta Scientiarum Naturalium Universitatis Pekinensis, 2023, 59(6): 1043-1051.
[6]	ZHANG Ling, CAO Jian, ZHANG Yuan, FENG Shuo, WANG Yuan. Reinforcement Learning of Spiking Neural Network Based on Knowledge Distillation [J]. Acta Scientiarum Naturalium Universitatis Pekinensis, 2023, 59(5): 757-763.
[7]	LIU Xiangcheng, CAO Jian, YAO Hongyi, XU Pengtao, ZHANG Yuan, WANG Yuan. AdaPruner: Adaptive Channel Pruning and Effective Weights Inheritance [J]. Acta Scientiarum Naturalium Universitatis Pekinensis, 2023, 59(5): 764-772.
[8]	WANG Nan, HAO Jinmin, HOSHINO Satoshi, TIAN Yufu. Study on the Relationship Between Industrial Structure and Land Use Structure of Rural Areas in Huang-Huai-Hai Plain: A Case Study of Quzhou County [J]. Acta Scientiarum Naturalium Universitatis Pekinensis, 2023, 59(5): 801-812.
[9]	ZHANG Chi, GUAN Ping, ZHANG Jihua, LIANG Xiaowei, DING Xiaonan, YOU Yuan. A Review of the Progress on Fractal Theory to Characterize the Pore Structure of Unconventional Oil and Gas Reservoirs [J]. Acta Scientiarum Naturalium Universitatis Pekinensis, 2023, 59(5): 897-908.
[10]	ZHANG Lu, PENG Yan, LI Qianhui, ZHANG Hongsheng, HE Qing, Ali Mamtimin. Characteristics and Dissimilarity of Turbulent Transport of Heat and Momentum in Summer Unstable Atmospheric Surface Layer in Taklimakan Desert and Its Physical Mechanisms [J]. Acta Scientiarum Naturalium Universitatis Pekinensis, 2023, 59(4): 581-592.
[11]	ZHOU Rong, WU Chaodong, ZHANG Yanan. Outpainting Reconstruction of Sandstone Thin-Section Image Based on Generative Adversarial Network [J]. Acta Scientiarum Naturalium Universitatis Pekinensis, 2023, 59(2): 231-241.
[12]	ZHOU Xiangyu, MAO Shanjun, LI Mei. Bearing Fault Diagnosis Method Based on Down-Sampling in Frequency Domain and CNN [J]. Acta Scientiarum Naturalium Universitatis Pekinensis, 2023, 59(2): 251-260.
[13]	ZHAO Dandan, ZHANG Junpeng, MENG Jiana, ZHANG Zhihao, SU Wen. Medical Entity Relation Extraction Based on Pre-trained Model and Hybrid Neural Network [J]. Acta Scientiarum Naturalium Universitatis Pekinensis, 2023, 59(1): 65-75.
[14]	XIE Hao, CAO Jian, LI Pu, ZHAO Xiongbo, ZHANG Xing. A Hardware Accelerator for SSD Object Detection Algorithm Based on FPGA [J]. Acta Scientiarum Naturalium Universitatis Pekinensis, 2022, 58(6): 1015-1022.
[15]	XU Pengtao, CAO Jian, SUN Wenyu, LI Pu, WANG Yuan, ZHANG Xing. Layer Pruning via Fusible Residual Convolutional Block for Deep Neural Networks [J]. Acta Scientiarum Naturalium Universitatis Pekinensis, 2022, 58(5): 801-807.

Hybrid Neural Network for Recognition of the “de” Structure with Semantic Ellipsis

基于组合神经网络的语义省略“的”字结构识别

RichHTML

PDF

Knowledge

Abstract

Cite this article

share this article

References

Related Articles 15

Recommended Articles

Metrics