一体化应急应战体系能力评估关键指标提取技术框架研究

Extraction Technical Framework for Capabilities Evaluation of the Integrated War Emergency System

面向一体化应急应战体系能力评估需求,传统基于指标体系构建聚合的评估方法具有指标提取困难、主观性强、海量动态数据难以描述等限制。提出一种数据驱动的体系评估技术框架结构,通过融合知识图谱和事理图谱两类技术,实现体系静态知识和动态知识的全面描述;通过实体、事件、关系等图谱要素数据自动抽取,实现体系框架的高效构建和持续更新;通过提供热点主题发现跟踪和态势评估预测等技术设计,实现可管理、可计算、可解释的一体化应急应战体系能力评估;通过一个有限场景的仿真实验,验证了本技术框架的可行性。

Concerning capabilities evaluation of the integrated war emergency system(IWES),the metrics system-based methods suffer from the limitation of metric acquisition difficulty,biased decision subjectivity,inability of describing massive dynamic data,etc.This paper proposes a new data-driven capabilities evaluation framework,which integrates the knowledge graph and event evolutionary graph to describe the domain static knowledge and dynamic events respectively.By synthesizing the automatic extraction techniques of entities,events,and relations,the framework can be built and updated efficiently.The framework also designs the application techniques of hot topic recognition and tracking,as well as event situation awareness and prediction.These characters enable the framework to solve the problem of IWES capabilities evaluation in a data manageable,computable,and explainable manner.Finally,a small-sized experiment based on simulated data sets is reported to validate the feasibility of the framework.