地球信息科学学报 ›› 2019, Vol. 21 ›› Issue (8): 1143-1151.doi: 10.12082/dqxxkx.2019.180654

• 地球信息科学理论与方法 •    下一篇

基于BiLSTM-CRF的中文层级地址分词

程博1,李卫红1,*(),童昊昕2   

  1. 1. 华南师范大学地理科学学院,广州 510631
    2. 航天精一(广东)信息科技有限公司,广州 510665
  • 收稿日期:2018-12-13 修回日期:2019-04-29 出版日期:2019-08-25 发布日期:2019-08-25
  • 通讯作者: 李卫红 E-mail:hongweili9981@163.com
  • 作者简介:程 博(1993-),男,湖北黄石人,硕士生,主要从事时空数据挖掘、自然语言处理。E- mail: <email>15625110283@163.com</email>
  • 基金资助:
    广东省重大科技专项(2017B030305005)

Chinese Address Segmentation based on BiLSTM-CRF

CHENG Bo1,LI Weihong1,*(),TONG Haoxin2   

  1. 1. South China Normal University, Guangzhou 510631, China
    2. Guangdong Finest Planning Information Technology Limited Company, Guangzhou 510665, China
  • Received:2018-12-13 Revised:2019-04-29 Online:2019-08-25 Published:2019-08-25
  • Contact: LI Weihong E-mail:hongweili9981@163.com
  • Supported by:
    Major Science and Technology Projects of Guangdong Province,(2017B030305005)

摘要:

中文地址分词是中文地址标准化的基础工作和地理编码的重要手段,同时也是中文分词和地理研究领域中关注的热点问题之一。针对当前中文地址分词方法缺乏地址层级切分和过多依赖词典和特征的问题,本研究结合四词位标注集和中文层级地址特点,构建针对中文层级地址分词的地址标注体系,并提出融合双向长短时记忆网络和条件随机场(BiLSTM-CRF)的中文层级地址分词模型。该模型既考虑了BiLSTM模型能够记忆上下文地址的特性,也保留了CRF算法可以通过转移概率矩阵控制地址标注输出的能力。针对该地址标注体系标注的训练地址样本,分别使用CRF、LSTM、BiLSTM与BiLSTM-CRF模型进行训练对比。结果表明:① 基于中文地址标注体系的模型分词效果更佳,地址标注更为精细,符合实际地址分布情况;② BiLSTM-CRF模型精确度达到93.4%,高于CRF(90.4%)、LSTM(89.3%)和BiLSTM(91.2%),其整体地址分词性能和各层级地址分词效果相对于其他模型更突出;③ 各模型分词性能与地址层级保持一致,即地址层级越高,分词效果越好。本研究提出的中文地址标注体系和分词模型为开展中文地址标准化工作提供了方法参考,同时也为进一步提升地理编码技术的精准度提供了可能。

关键词: 中文分词, 地址标注, 中文层级地址分词, 长短时记忆网络(LSTM), 双向长短时记忆和条件随机场模型(BiLSTM-CRF)

Abstract:

Chinese word segmentation is a basic step in Chinese text processing and Chinese natural language processing. As a branch of Chinese word segmentation, Chinese address segmentation, which has become one of the hottest issues in Chinese word segmentation and geography research, is an important method to standardize Chinese address and conduct geocoding. Existing studies on Chinese address segmentation are mainly based on Statistical Machine Learning (SML) and Recurrent Neural Network (RNN). However, either of them cannot combine the advantages of the other. Furthermore, the current Chinese address segmentation methods lack address level segmentation and are too dependent on dictionaries and features. Therefore, this paper combined the four-word-position tagging set and Chinese hierarchical address characteristics to construct an address tagging system, and proposed a Chinese hierarchical address segmentation model (BiLSTM-CRF) which combines bidirectional long-term memory networks and conditional random fields algorithm. The proposed model utilizes the BiLSTM model to remember the characteristics of context address, while retaining the ability of the CRF algorithm to control the address tagging output by transferring probability matrix. In so doing, it has more powerful capability than the traditional statistical machine learning algorithms and RNN in the field of sequence labeling and word segmentation. To test the performance of BiLSTM-CRF on address samples marked by the address tagging system, CRF, LSTM, and BiLSTM were used to compare with BiLSTM-CRF and were respectively applied for training under the same condition as BiLSTM-CRF. We found that: (1) The segmentation effect of BiLSTM-CRF which is based on the Chinese address tagging system was better than the models for comparison, and the address tagging was more elaborate, in line with the actual address distribution. (2) The BiLSTM-CRF model had an accuracy of 93.4%, which was higher than the CRF (90.4%), LSTM (89.3%) and BiLSTM (91.2%) models. The overall address word segmentation performance and the effect of BiLSTM-CRF on each level address segmentation were more prominent than the other models. (3) The word segmentation performance of each model was correlated with the address level positively, i.e., the higher the address level, the better the word segmentation effect. The Chinese address tagging system and word segmentation model proposed in this study give a reference for the standardization of Chinese address, and provide the possibility to further improve the accuracy of geocoding technology. Future study can focus on fine-tuning the model to improve the model accuracy.

Key words: Chinese word segmentation, address tagging, hierarchical word segmentation, Long Short-Term Memory (LSTM), Bidirectional Long Short-Term Memory-Conditional Random Field (BiLSTM-CRF)