测绘通报 ›› 2022, Vol. 0 ›› Issue (5): 120-125.doi: 10.13474/j.cnki.11-2246.2022.0152

• 学术研究 • 上一篇    下一篇

结合地名知识和位置信息的中文地名匹配方法

姜松言1, 宋国民1, 贾奋励1, 陈令羽1, 张蓝天2   

  1. 1. 信息工程大学, 河南 郑州 450001;
    2. 北京市遥感信息研究所, 北京 100192
  • 收稿日期:2022-01-17 发布日期:2022-06-08
  • 通讯作者: 宋国民。E-mail:imsonny@foxmail.com
  • 作者简介:姜松言(1997-),男,硕士生,研究方向为地名数据融合与处理。E-mail:jiangsongyan1997@163.com
  • 基金资助:
    国家自然科学基金(42171456);国家重点研发计划(2017YFB0503503);国家自然科学基金青年基金(41801317)

Combining toponymy knowledge and location information for Chinese place names matching

JIANG Songyan1, SONG Guomin1, JIA Fenli1, CHEN Lingyu1, ZHANG Lantian2   

  1. 1. Information Engineering University, Zhengzhou 450001, China;
    2. Beijing Institute of Remote Sensing Information, Beijing 100192, China
  • Received:2022-01-17 Published:2022-06-08

摘要: 针对中文地名数据中存在的“同名异地”“异地同名”等歧义问题,本文提出了一种结合地名知识和位置信息的地名匹配方法。首先基于特征词的规则匹配模式对地名名称结构进行了解析。然后结合地名的位置信息实现中文地名匹配;同时,基于地名的行政区划信息和实体信息关联多源地名数据,构建地名知识库。最后以兴趣点数据为例,基于名称相似度和位置相似度匹配地名数据中的地名信息,并以知识图谱的形式进行存储和表达。

关键词: 地名知识, 地名匹配, 相似度计算, 数据消歧

Abstract: Aiming at the ambiguity in Chinese place names such as "same name for different locations" and "different names for same location", we propose a method for place names matching with combining toponymy knowledge and location information. Firstly, the semantic structure of place name is analyzed based on rule matching pattern of feature words. Afterwards, combining the location information, place names matching is achieving. Meanwhile, place name data is correlated by the entity information and administrative division to construct toponymy knowledge base. Taking POI data as instance, we implement place names matching with name similarity and location similarity, and the results are stored and represented in the form of knowledge graph.

Key words: toponymy knowledge, place names matching, similarity calculation, data disambiguation

中图分类号: