面向灾害预警的空间数据概念增强

Spatial data concept enhancement for disaster warning

现阶段,灾害应急的空间数据为多源异构数据,存在质量良莠不齐、语义模型不统一、数学基础和分辨率不明确等问题。为了提高应急数据处理效率,有必要对现有空间数据进行概念增强处理。在研究了多种空间数据结构之后,本文提出了建立通用元数据模型以进行高层次的结构与模式匹配,同时建立空间数据概念增强范式,以此对空间地物中单目标特征度量、目标间关系识别、群目标重要性排序、群目标聚类分析、群目标结构关系识别进行研究,提取灾害信息数据并对其进行本体建模,组合形成数据概念增强智能体。实验结果表明,此智能体可挖掘灾害隐藏信息,提高预警效率,实现可视化的空间数据概念增强。

At present,most emergence data is multi-source heterogeneous spatial data,and exists some problems such as uneven data quality,inconsistent semantic model,unclear mathematical foundation and resolution.After studying various spatial data structures,this paper proposes a proper general metadata model for high-level structure and pattern matching,and establishes a spatial data concept enhancement paradigm to perform research on single-objective feature measurement,relation detection between objects,group object importance ranking,group object cluster analysis,group object structure relationship identification.And the disaster information is extracted and modeled using ontology to form the data enhancement agent.The experiment results show that this agent can mine disaster hidden information,improve early warning efficiency,and enhance the concept of visual spatial data.