大数据时代细胞仿生结构的地理空间认知方法

Geospatial cognition method of biomimetic cell structure under the era of big data

针对多属性、多维、多源异构的时空大数据特征,急需一种全空间的时空大数据模型进行地理空间认知。基于地理空间认知过程,提出了地理空间细胞模型,将地理空间抽象成无数个地理细胞组成,邻近的地理细胞围绕核心细胞聚集为地理簇,相似属性的地理簇聚集为地理块,实现了时空大数据的高效聚类,解决了时空大数据的组织管理问题,达到了时空大数据在地理空间中微观与宏观的统一认知。通过仿生生物遗传学里的细胞、组织、器官、系统管理地理空间细胞的聚类过程,形成对地理空间自下而上的多尺度表达。

As China attaches great importance to the development of big data industry,efficient clustering of spatio-temporal big data and mining of geospatial information becomes the focus of research.According to the spatio-temporal big data characteristics of multi-attribute,multi-dimensional,and multi-source and heterogeneity,a full space spatio-temporal big data model is urgently needed for geospatial cognition.In this paper,a cell model of geographical space is proposed based on the cognitive process of geographical space.Geographical Space is abstracted into numerous geographical cells,which are clustered into geographical clusters around the core cells.The geographic clustering of similar attributes can be divided into geographic blocks,which can realize the efficient clustering of spatio-temporal data,solve the organization and management of spatio-temporal data,and achieve the unified cognition of spatio-temporal big data in micro and macro in geospatial.Clustering processes of geospatial cells are managed through simulating cells,tissues,organs,and systems in biogenetics,to form a multi-scale representation of the bottom-up.