论文部分内容阅读
企业在事故隐患排查治理过程中积累了大量隐患数据,为挖掘其潜在价值,实现事故隐患预警预控,针对隐患类型多、数量大的特点,应用垂直数据格式挖掘算法对高维隐患数据进行关联规则挖掘,并利用Kulc和不平衡比(IR)减小隐患出现频率差异对规则的影响;在此基础上,设计基于关联规则的隐患预警评估模型,并对预警信息进行可视化处理,最终构建完整的企业隐患预警方法。以130家机械制造企业的53 029条隐患数据为例,验证所建预警方法的可行性。结果表明,该方法对事故隐患预警的准确率为80.62%。
In order to excavate the potential value of hidden dangers, enterprises have accumulated a large amount of hidden data in the process of investigation and management of hidden dangers, and realized the early warning and pre-control of hidden dangers. According to the characteristics of many types and large quantities of hidden dangers, vertical data format mining algorithms are used to correlate high-dimensional hidden data Rule mining, and using Kulc and the imbalance ratio (IR) to reduce the impact of the frequency of hidden dangers on the rules; on this basis, the design of risk warning assessment model based on association rules, and early warning information visual processing, and ultimately build a complete Business risk warning method. Taking 130 029 hidden data of 130 machinery manufacturing enterprises as an example, the feasibility of the built early warning method is verified. The results show that the accuracy of the method for early warning of hidden accidents is 80.62%.