A Comparative Study of Chinese Address Segmentation Methods
Authors
Jiaqi Yu*
College of Remote Sensing and Technology, Wuhan University, Wuhan, Hubei Province, China, 430072
*Corresponding author. Email: 2019302130161@whu.edu.cn
Corresponding Author
Jiaqi Yu
Available Online 4 July 2022.
- DOI
- 10.2991/assehr.k.220701.038How to use a DOI?
- Keywords
- word segmentation; natural language processing; Chinese; deep learning
- Abstract
Nowadays, natural language processing continues to grow with its popularity in research and commercial fields. With this trend happening, researchers now put more effort into applying machine learning to achieve natural language processing. This paper concentrates on the word segmentation aspect of Chinese natural language processing, and introduces and compares Bi-LSTM-CRF model and typical toolkits for Chinese word segmentation, aiming for a better understanding of which method to choose on a limited training basis. It can be carried out that when training at a small dataset scale, Bi-LSTM-CRF model segments addresses more accurately than typical toolkits.
- Copyright
- © 2022 The Authors. Published by Atlantis Press SARL.
- Open Access
- This is an open access article distributed under the CC BY-NC 4.0 license.
Cite this article
TY - CONF AU - Jiaqi Yu PY - 2022 DA - 2022/07/04 TI - A Comparative Study of Chinese Address Segmentation Methods BT - Proceedings of the 2022 International Conference on Science and Technology Ethics and Human Future (STEHF 2022) PB - Atlantis Press SP - 193 EP - 196 SN - 2352-5398 UR - https://doi.org/10.2991/assehr.k.220701.038 DO - 10.2991/assehr.k.220701.038 ID - Yu2022 ER -