论文部分内容阅读
应急管理研究的深入,现有知识表示方法难以满足其跨学科、知识异构的特点。知识元模型的提出为解决这一现状提供了可能。针对知识元模型的特征,以短语结构树为基础,提出一种基于规则的知识元属性抽取方法。以搜集的2000至2009年应急管理矿难案例为数据源进行实验,并对知识元属性抽取结果进行评估与分析。结果表明,该方法能基本满足从大规模数据中自动抽取知识元及属性,将属性抽取方法应用到应急管理中,提高了知识元抽取效率。
In-depth research of emergency management, existing knowledge representation method is difficult to meet its interdisciplinary, knowledge heterogeneity characteristics. The proposed meta-model of knowledge provides the possibility to solve this situation. Aiming at the characteristics of knowledge meta-model, a phrase-based knowledge element attribute extraction method based on the phrase tree is proposed. Experiments were conducted on the collected data of emergency management mining cases from 2000 to 2009, and the results of attribute extraction of knowledge elements were evaluated and analyzed. The results show that this method can basically satisfy the automatic extraction of knowledge elements and attributes from large-scale data, and apply the attribute extraction method to emergency management, which improves the efficiency of knowledge element extraction.