基于基因结构的复杂应急案例表示方法

发布时间：2019-04-17 05:56

【摘要】：案例推理作为人工智能领域类比推理的有效方法,其检索效果与推理能力很大程度上依赖于案例构建的合理与否.考虑到已有方法在表示复杂应急案例上的不足,本文依据历史案例将复杂突发事件的"情境-情景"结构抽象化,以此提出复杂应急案例基因模型的概念.基于基因模型构建多级的案例结构,能形成可重复使用的、标准的、简化的、可供应急决策者重组参考且具有"生物型"特征的复杂应急案例表示模型.该方法在一定程度上继承与集成了框架与本体表示法的特点,实现了复杂应急案例结构的系统性表达,使案例内容有机连接成整体,可更好地服务于复杂应急案例推理.为更好地说明,本文以"台风/暴雨"引发的复杂情境链为例,阐述基因结构表示复杂应急案例的合理性与有效性.
[Abstract]:Case-based reasoning (CBR), as an effective method of analogy reasoning in artificial intelligence, its retrieval effect and reasoning ability depend on the rationality of case-building to a great extent. Considering the deficiency of existing methods in expressing complex emergency cases, this paper abstracts the "situation-context" structure of complex emergencies according to historical cases, and then puts forward the concept of genetic model of complex emergency cases. The multi-level case structure based on gene model can form a complex emergency case representation model, which can be reused, standardized and simplified, which can be used as a reference for emergency decision makers and has the characteristics of "biotype". To a certain extent, this method inherits and integrates the characteristics of framework and ontology representation, realizes the systematic expression of complex emergency case structure, and makes the case content organically connected to the whole, which can better serve the complex emergency case-based reasoning. In order to better illustrate the rationality and effectiveness of gene structure in representing complex emergency cases, the complex situation chain caused by typhoon / rainstorm is taken as an example in this paper.
【作者单位】：哈尔滨工业大学经济与管理学院;
【基金】：国家自然科学基金重大研究计划项目(91024028,91024031,91324018);国家自然科学基金面上项目(71473058)~~
【分类号】：D63

【相似文献】