地理科学

地理科学 ›› 2016, Vol. 36 ›› Issue (8): 1180-1189.doi: 10.13249/j.cnki.sgs.2016.08.008

• • 上一篇    下一篇

地理空间元数据关联网络的构建

赵红伟1,2,3(), 诸云强1,2, 侯志伟1,2,3, 杨宏伟4   

  1. 1. 中国科学院资源与环境信息系统国家重点实验室, 北京100101
    2. 中国科学院地理科学与资源研究所,北京100101
    3. 中国科学院大学, 北京 100049
    4. 中国石油规划总院, 北京100000
  • 收稿日期:2015-11-23 修回日期:2016-05-04 出版日期:2016-08-20 发布日期:2016-08-20
  • 作者简介:

    作者简介：赵红伟（1987-）,女,山东聊城人,博士研究生,主要研究方向为地理空间数据语义关联、地理空间数据共享。E-mail:zhaohw.10s@igsnrr.ac.cn

  • 基金资助:
    国家自然科学基金项目（41371381）、科技部科技基础性工作专项项目（2013FY110900）、国家重大科学仪器设备开发专项（2012YQ06002704）、云南省科技计划项目（2012CA021）资助

Construction of Geospatial Metadata Association Network

Hongwei Zhao1,2,3(), Yunqiang Zhu1,2, Zhiwei Hou1,2,3, Hongwei Yang4   

  1. 1. State Key Laboratory of Resources and Environmental Information System, Chinese Academy of Sciences, Beijing 100101, China
    2. Institute of Geographic Sciences and Natural Resources Research, Chinese Academy of Sciences, Beijing 100101, China
    3. University of Chinese Academy of Sciences, Beijing 100049, China
    4.China Petroleum Planning & Engineering Institute, Beijing 100000, China
  • Received:2015-11-23 Revised:2016-05-04 Online:2016-08-20 Published:2016-08-20
  • Supported by:
    National Nature Science Foundation of China (41371381), Science and Technology Basic Work of Science and Technology (2013FY110900), the National Key Scientific Instrument and Equipment Development Project (2012YQ06002704), Science and Technology Project of Yunnan Province (2012CA021)

RichHTML

PDF (PC)

赞

被引次数

12 | 4

摘要：

利用资源描述框架（RDF）设计地理空间元数据关联模型,根据地理空间元数据之间的语义关系和语义相关度的计算,以构建以元数据为节点、元数据之间的语义关系为边、语义相关度为权重的关联网络。在这一网络中,一个节点是一个地理空间元数据的资源描述图,包含属性特征（数据来源、空间特征、时间特征、内容）及其关系特征（元数据之间的语义关系、语义相关度）。实验及其分析表明,地理空间元数据关联网络可以有效地支持地理空间数据语义关联检索、推荐等应用,这与传统的基于关键词的元数据检索方式相比,具有更高的准确度。

关键词: 地理空间元数据, 关联数据, 语义相似度, 关联网络

Abstract:

The rapid acquisition of geospatial data mainly depends on geospatial metadata. But the traditional organization of geospatial metadata and the keywords-based retrieval methods create barriers among metadata considering semantic relations between geospatial data such as spatial topology relationship, category relationship, resulting in a bottleneck in geospatial data sharing. In the context of big geospatial data, the development of linked data provides an effective practice for the semantic sharing and application of massive geospatial data. The linked geodata is intended to break the semantic barriers between geospatial data and form a data network with semantic realtions. Due to the complexity, diversity and uncertainy of geospatial data, linked geodata is often achieved through the association between metadata. Geospatial metadata contains a number of descriptive information. How to effectively organize vast amounts of geospatial metadata and map the metadata into the semantic space by simple way have become the hotspots in the field of geospatial data sharing. Construction of semantic associations among geospatial metadata is an effective means of performing semantic retrieval using related data technologies. Effective application of linked data depends on effective association models. Considering this, a method of constructing geospatial metadata association networks is proposed in this paper: firstly, a geospatial metadata association model is designed on basis of the resource description framework (RDF); secondly, a semantic relation between metadata is determined and the relationship is constructed; and finally, the degree of semantic relevance of the semantic relationship is calculated. In the association network, the metadata are nodes, the semantic relationships between the metadata are edges, and the degrees of semantic relevance are the weights of the edges. Every node is an RDF that has attribute properties, such as sources, spatial characteristics, temporal characteristics, and content, and has properties of semantic relationships. Experimental results showed that the constructed network could effectively support operations such as semantic association search and recommendation, and the retrieval results were more precise and accurate compared with traditional metadata retrieval methods based on keywords.

Key words: geospatial metadata, linked data, semantic relevance, association network

中图分类号: 

  • TP391