| 基于随机森林的突发事件分级模型研究 |
| |
| 引用本文： | 徐绪堪,王京. 基于随机森林的突发事件分级模型研究[J]. 中国安全生产科学技术, 2018, 14(2): 77-81. DOI: 10.11731/j.issn.1673-193x.2018.02.012 |
| |
| 作者姓名： | 徐绪堪 王京 |
| |
| 作者单位： | （1.河海大学 企业管理学院，江苏 常州，213000；2.南京大学 信息管理学院，江苏 南京，210023） |
| |
| 摘 要： | 突发事件的分级是快速响应和有效应对的基础，为了解决目前突发事件分级宽泛、主观性强和动态适应性弱等问题，将多分类器集成引入突发事件的分级过程中，借助随机森林组合分类器，针对突发事件特征，构建突发事件分级的随机森林模型，形成突发事件分级过程，进而对事前的应急预案制定、事中的应急资源配置和应急决策提供有力支撑；最后，以2014-2016年洪涝灾害突发事件为例检验构建的模型和过程。研究结果表明:与支持向量机方法相比，通过检验随机森林组合分类器分类结果准确率达到97.56%，在突发事件分级的应用上是可行的，进而可为突发事件的快速响应和应急决策提供信息支撑和参考依据。
|
| 关 键 词： | 突发事件 分级 随机森林 |
Research on classification model of emergencies based on random forest |
| |
| XU Xukan1,2,WANG Jing1. Research on classification model of emergencies based on random forest[J]. Journal of Safety Science and Technology, 2018, 14(2): 77-81. DOI: 10.11731/j.issn.1673-193x.2018.02.012 |
| |
| Authors: | XU Xukan1 2 WANG Jing1 |
| |
| Affiliation: | (1. School of Business Administration, Hohai University, Changzhou Jiangsu 213000, China;2. School of Information Management, Nanjing University, Nanjing Jiangsu 210023, China) |
| |
| Abstract: | The classification of emergencies is the foundation of the rapid and effective response. To solve the problems that the current classification of emergencies is broad, the subjectivity is strong, and the dynamic adaptability is weak, the multiple classifiers were integrated into the classification process of emergencies, and the classifiers were combined by using the random forest. Aiming at the characteristics of emergencies, a random forest model for the classification of emergencies was constructed, then the classification process of emergencies was formed, so as to provide the support for the emergency plan formulation before the emergencies, and the emergency resource configuration and emergency decision-making during the emergencies. Finally, the constructed model and process were verified by taking the emergencies of flood disasters from 2014 to 2016 as example. The results showed that compared with the support vector machine (SVM) method, the accuracy of the classification results by the combined classifiers of random forest was 97.56%. It is feasible and effective in the application for the classification of emergencies, and can provide the information support and reference basis for the rapid response and emergency decision-making of emergencies. |
| |
| Keywords: | emergencies classification random forest |
| 本文献已被 CNKI 等数据库收录！ |
| 点击此处可从《中国安全生产科学技术》浏览原始摘要信息 |
|
点击此处可从《中国安全生产科学技术》下载免费的PDF全文 |
|
