期刊论文

Causation network analysis of the construction falling or collapsing accidents based on the text mining

中文

安全与环境学报

1009-6094

2020

20

04

1284-1290

10.13637/j.issn.1009-6094.2019.0497

71371036:国家自然科学基金 CX20190669:湖南省研究生科研创新项目

本校为第一机构

为明确建筑施工过程中高处坠落事故的致因及其之间的联系,选取2012—2018年国内558例建筑施工高处坠落事故调查报告,借助文本挖掘和社会网络分析的方法,对事故报告进行分词处理、特征选择后,得到致因特征项、致因网络及致因集合,并利用词云和网络结构图等方法对结果进行可视化展示。结果表明,文本挖掘所得的28项致因中,7项致因位于致因网络的核心区域,基于该7项致因分别构建了致因集合,为建筑施工高处坠落事故的预控提供参考。

The present paper has collected 558 reports in investigating the construction fall-from-height accidents that have occurred from 2012 to 2018 in China.In the paper,we have cherished our goal and inclinations to identify and clarify the root causes of such falling accidents in the construction due to the internal multiplicity of correlated factors among them as the research corpus in combination with the context exploration through social network analysis.On the basis of processing the corpus by R,we have carried out a text feature items dimensi...