主管单位:中国科学技术协会
主办单位:中国地理学会
承办单位:华东师范大学

世界地理研究 ›› 2020, Vol. 29 ›› Issue (1): 71-76.DOI: 10.3969/j.issn.1004-9479.2020.01.2019800

• 世界政治与经济 • 上一篇    下一篇

GDELT:感知全球社会动态的事件大数据

沈石1,2,3(), 宋长青1,2,3(), 程昌秀1,2,3, 高剑波2,3, 叶思菁1,2,3   

  1. 1.北京师范大学地表过程与资源生态国家重点实验室, 北京 100875
    2.北京师范大学地理科学学部, 北京100875
    3.北京师范大学地理数据与应用分析中心, 北京100875
  • 收稿日期:2019-09-10 修回日期:2019-12-21 出版日期:2020-01-20 发布日期:2022-01-22
  • 通讯作者: 宋长青
  • 作者简介:沈石(1990-),男,讲师,博士,主要研究方向:地理数据挖掘与分析,E-mail:shens@bnu.edu.cn
  • 基金资助:
    第二次青藏高原综合考察研究资助(2019QZKK0608)

GDELT: Big event data for sensing global social dynamics

Shi SHEN1,2,3(), Changqing SONG1,2,3(), Changxiu CHENG1,2,3, Jianbo GAO2,3, Sijing YE1,2,3   

  1. 1.State Key Laboratory of Earth Surface Processes and Resource Ecology, Beijing Normal University, Beijing 100875, China
    2.Faculty of Geographical Science, Beijing Normal University, Beijing 100875, China
    3.Center for Geodata and Analysis, Beijing Normal University, Beijing 100875, China
  • Received:2019-09-10 Revised:2019-12-21 Online:2020-01-20 Published:2022-01-22
  • Contact: Changqing SONG

摘要:

正确解析国家间政治关系及其演化过程是开展地缘关系研究的重要基础。从大数据的角度开展地缘关系研究为该领域的探索提供了一种新的途径。国家或区域的局部政治倾向数据无法为地缘关系研究提供全面和翔实的数据支撑。论文介绍的一个全新的事件数据库GDELT(Global Database of Event, Language , Tone),它在诸多方面弥补了传统数据的不足。该数据不仅详细记录了全球范围事件的发生时间、地点、内容以及参与者信息,而且系统地对事件进行分类和评分。本文从数据内容、事件评分和分类体系三方面详细介绍GDELT数据,并总结了该数据的优势和潜在研究方向,以期为我国地缘关系研究等领域提供帮助和参考。

关键词: GDELT, 地缘关系, 事件数据, 戈尔德斯坦量表

Abstract:

Properly analyzing the international political relations and their evolution process is an essential for geo-relationships research. Geo-relationships research from the perspective of big data provides a new approach for the exploration in this field. Data reflecting international political trends by traditional methods is unable to provide comprehensive and informative data support for regional or global research. A new event database GDELT (Global Database of Event, Language, Tone) introduced in this article has offset the deficiency of traditional data in many aspects. The data not only records the date-time, place, content and participant of events worldwide but also systematically classifies and scores events. This article introduces GDELT data in detail from three aspects of data content, event scoring, and classification system, and summarize the advantages and potential research directions of the data, intending to provide help and reference for China’s geo-relationships research and other fields.

Key words: GDELT, geo-relationships, event data, Goldstein scale