文本中空间位置识别技术的研究进展  

Research progress on spatial location recognition technologies in text

作　　者：陈紫怡 李大军[1] 谭永滨[1] CHEN Ziyi;LI Dajun;TAN Yongbin(School of Surveying and Geo Information Engineering,East China University of Technology,Nanchang,Jiangxi 330013,China)

机构地区：[1]东华理工大学测绘与空间信息工程学院,江西南昌330013

出　　处：《北京测绘》2025年第2期127-134,共8页Beijing Surveying and Mapping

基　　金：国家自然科学基金(42261078);江西科技厅项目(20223BBE51030)。

摘　　要：大数据时代的到来给地理信息行业带来了新的挑战,在文本爆炸式增长的背景下,提取其中的空间位置对事件分析、社会感知及灾难响应等具有重要意义。提取文本中的空间位置分为两个步骤:识别与定位。本文从这两个步骤出发,总结了地名实体识别技术的发展阶段及优缺点,介绍了地名定位过程所涉及的三大关键技术方法:地名数据库的融合技术、地名匹配技术以及消歧技术。最后讨论了现阶段提取空间位置存在的问题和挑战。The advent of the big data era has brought new challenges to the geographic information industry.Against the backdrop of explosive growth in text data,extracting spatial locations from the data is of great significance for event analysis,social perception,and disaster response.The extraction of spatial locations from text involves two main steps:identification and location.Starting from these steps,this paper summarized the development stages and pros and cons of geographic name entity recognition technology.It introduced three key technical methods involved in the process of geographic name location:fusion technology of geographic name databases,geographic name matching technology,and disambiguation technology.Finally,it discussed the current problems and challenges in extracting spatial locations.

关 键 词：地名实体识别 深度学习 自然语言处理 定位 相似度匹配 

分 类 号：P208[天文地球—地图制图学与地理信息工程]