Cross-domain Chinese Word Segmentation Based on New Word Discovery

ZHANG Jun; LAI Zhipeng; LI Xue; NING Gengxin; YANG Cui

doi:10.11999/JEIT210675

Volume 44 Issue 9

Sep. 2022

Turn off MathJax

Article Contents

Article Navigation > Journal of Electronics & Information Technology > 2022 > 44(9): 3241-3248

ZHANG Jun, LAI Zhipeng, LI Xue, NING Gengxin, YANG Cui. Cross-domain Chinese Word Segmentation Based on New Word Discovery[J]. Journal of Electronics & Information Technology, 2022, 44(9): 3241-3248. doi: 10.11999/JEIT210675

Citation:

ZHANG Jun, LAI Zhipeng, LI Xue, NING Gengxin, YANG Cui. Cross-domain Chinese Word Segmentation Based on New Word Discovery[J]. Journal of Electronics & Information Technology, 2022, 44(9): 3241-3248. doi: 10.11999/JEIT210675

Citation:

PDF( 619 KB)

Cross-domain Chinese Word Segmentation Based on New Word Discovery

doi: 10.11999/JEIT210675

School of Electronic and Information Engineering, South China University of Technology, Guangzhou 510641, China

Funds: The National Natural Science Foundation of China (61871191), The Natural Science Foundation of Guangdong Province (2020A1515010962), The Natural Science Foundation of Guangzhou (202002030251)

Received Date: 2021-07-06
Accepted Date: 2021-09-14
Rev Recd Date: 2021-09-14

Available Online: 2021-12-25

Publish Date: 2022-09-19

Abstract

Abstract

Deep Neural Network (DNN) is the major method in current Chinese word segmentation. However, its performance is significantly degraded when the network trained for one domain is used in other domains due to the Out Of Vocabulary (OOV) words and expression gaps. In this paper, a cross domain Chinese word segmentation system based on new word discovery is built to handle the OOV word and expression gap problems. An unsupervised new word discovery algorithm based on vector enhanced mutual information and weighted adjacency entropy, and a Chinese word segmentation model based on adversarial training are also proposed to improve the performance of the baseline system. Experimental results show that the proposed method is superior to the conventional methods in the OOV rates, precisions, recalls and F-scores.
- Chinese word segmentation,
- New word discovery,
- Cross-domain,
- Vector enhancement mutual information,
- Adversarial training

FullText(HTML)

References(16)

References

[1]	陈平, 刘晓霞, 李亚军. 基于字典和统计的分词方法[J]. 计算机工程与应用, 2008, 44(10): 144–146. doi: 10.3778/j.issn.1002-8331.2008.10.042 CHEN Ping, LIU Xiaoxia, and LI Yajun. Chinese word segmentation based on dictionary and statistics[J]. Computer Engineering and Applications, 2008, 44(10): 144–146. doi: 10.3778/j.issn.1002-8331.2008.10.042
[2]	WU Andi and JIANG Zixin. Word segmentation in sentence analysis[C]. 1998 International Conference on Chinese Information Processing, Beijing, China, 1998: 169–180.
[3]	朱聪慧, 赵铁军, 郑德权. 基于无向图序列标注模型的中文分词词性标注一体化系统[J]. 电子与信息学报, 2010, 32(3): 700–704. doi: 10.3724/SP.J.1146.2009.00214 ZHU Conghui, ZHAO Tiejun, and ZHENG Dequan. Joint Chinese word segmentation and POS tagging system with undirected graphical models[J]. Journal of Electronics &Information Technology, 2010, 32(3): 700–704. doi: 10.3724/SP.J.1146.2009.00214
[4]	YUAN Zheng, LIU Yuanhao, YIN Qiuyang, et al. Unsupervised multi-granular Chinese word segmentation and term discovery via graph partition[J]. Journal of Biomedical Informatics, 2020, 110: 103542. doi: 10.1016/j.jbi.2020.103542
[5]	DU Jinlian, MI Wei, and DU Xiaolin. Chinese word segmentation in electronic medical record text via graph neural network-bidirectional LSTM-CRF model[C]. 2020 IEEE International Conference on Bioinformatics and Biomedicine, Seoul, Korea, 2020: 985–989.
[6]	WANG Qi, ZHOU Yangming, RUAN Tong, et al. Incorporating dictionaries into deep neural networks for the Chinese clinical named entity recognition[J]. Journal of Biomedical Informatics, 2019, 92: 103133. doi: 10.1016/j.jbi.2019.103133
[7]	XU Jingjing, MA Shuming, ZHANG Yi, et al. Transfer deep learning for low-resource Chinese word segmentation with a novel neural network[C]. The 6th National CCF Conference on Natural Language Processing and Chinese Computing, Dalian, China, 2017: 721–730.
[8]	BELLEGARDA J R. Statistical language model adaptation: Review and perspectives[J]. Speech Communication, 2004, 42(1): 93–108. doi: 10.1016/j.specom.2003.08.002
[9]	刘伟童, 刘培玉, 刘文锋, 等. 基于互信息和邻接熵的新词发现算法[J]. 计算机应用研究, 2019, 36(5): 1293–1296. doi: 10.19734/j.issn.1001-3695.2017.11.0745 LIU Weitong, LIU Peiyu, LIU Wenfeng, et al. New word discovery algorithm based on mutual information and branch entropy[J]. Application Research of Computers, 2019, 36(5): 1293–1296. doi: 10.19734/j.issn.1001-3695.2017.11.0745
[10]	罗桂琼, 费洪晓, 戴弋. 基于反序词典的中文分词技术研究[J]. 计算机技术与发展, 2008, 18(1): 80–83. LUO Guiqiong, FEI Hongxiao, and DAI Yi. Research of Chinese segmentation based on converse segmentation dictionary[J]. Computer Technology and Development, 2008, 18(1): 80–83.
[11]	YAO Yushi and HUANG Zheng. Bi-directional LSTM recurrent neural network for Chinese word segmentation[C]. The 23rd International Conference on Neural Information Processing, Kyoto, Japan, 2016: 345–353.
[12]	LIU Liyuan, SHANG Jingbo, REN Xiang, et al. Empower sequence labeling with task-aware neural language model[C]. The Thirty-Second AAAI Conference on Artificial Intelligence, New Orleans, United States, 2018.
[13]	KAN Zhigang, QIAO Linbo, YANG Sen, et al. Event arguments extraction via dilate gated convolutional neural network with enhanced local features[J]. IEEE Access, 2020, 8: 123483–123491. doi: 10.1109/ACCESS.2020.3004378
[14]	MIKOLOV T, CHEN Kai, CORRADO G, et al. Efficient estimation of word representations in vector space[C]. The 1st International Conference on Learning Representations, Scottsdale, Arizona, 2013.
[15]	KIM Y. Convolutional neural networks for sentence classification[C]. The 2014 Conference on Empirical Methods in Natural Language Processing, Doha, Qatar, 2014: 1746–1751.
[16]	Beijing Universty, City University of Hong Kong, CKIP, et al. The second international Chinese word segmentation bakeoff data[EB/OL]. http://sighan.cs.uchicago.edu/bakeoff2005/, 2005.