时空大数据时代的地理知识工程展望

伴随智慧城市的逐步建设，日益成熟的传感网技术使时空数据的获取变得容易。遥感影像、地图瓦片、视频监控、网络爬虫、社交平台等各类时空数据常有PB级的巨量数据存储，人类社会已全面进入时空大数据时代。然而，海量数据的骤增与信息增值机制的缺乏形成了鲜明的对比，仅仅依靠信息技术尚难以快速地产生地理知识。如何利用新兴的信息化技术手段，结合地理信息及地理分析手段，快速挖掘、产生和利用时空数据成为当前亟待解决的关键问题。地理知识工程围绕人类空间认知与复杂知识发掘技术，对大数据时代的智能化应用和政府决策有重要的意义。立足于时空大数据背景，展开地理知识工程的内涵、时代特点、结构组成及研究方向等论述。地理知识工程的发展将在以虚拟地理环境、数字地球、智慧城市为代表的新型地理信息系统推动下，促进传统地理信息系统向着符合人类认知特点和探索过程的方向转变。通过海量时空数据到地理知识的快速转化，最终将解决"地理数据爆炸但地理知识贫乏"这一重要难题，提升数据的使用效能。

Prospect of Geo-Knowledge Engineering in the Era of Spatio-Temporal Big Data

With the construction of the smart city, the maturing sensor technology makes the spatio-temporal available. Every year, the amount of various kinds of spatio-temporal data from remote sensors, map servers, video monitoring, crawler and social platform reach up to PBs. Human society has entered the era of spatio-temporal big data. However, the sharp contrast between the sudden increase of mass data and the lack of information appreciation mechanism appears. The geo-knowledge produced by only IT are blacked seriously. How to use the emerging technology, combined with geographic information and accordingly analysis means, to produce, mine and use the spatio-temporal data has become the key problem to be solved urgently. Geo-knowledge is centered on human spatial cognition and complex knowledge discovery technology, which is of great significance for intelligent applications and government decision-making in the age of big data. In the era background of spatio-temporal big data, this paper discusses the connotation, time features, structure and research directions of geo-knowledge engineering in the future. Strongly driven by the virtual geographical environment, digital earth, smart city, geo-know-ledge engineering will promote the traditional GIS transforming to the directions of human cognition and exploration procedure. Through the rapid transformation from massive spatio-temporal data to the geo-knowledge, geo-knowledge engineering promises to solve the challenge problem of "geo-know-ledge explosion but knowledge poor" which will increase the efficiency of data use.