群体性突发事件特征要素的关联规则挖掘分析

Association Rules Mining and Analysis of Feature Elements of Group Emergencies

针对群体性突发事件预警防控存在的识别精度不高等问题,通过收集近年来国内典型群体性突发事件案例,在进行要素抽取和结构化定义的基础上,运用数据挖掘关联规则分析方法研究了事件发生时间、地点、诱发因素、参与人数等多类要素之间的关联性。研究结果表明,群体性突发事件的发生地点与发生日期及具体时间存在强关联关系,其中发生在政府和企业单位的群体性突发事件集中发生于工作日的下午;发生在公共场所区域的、以表达利益诉求为主要目的群体性突发事件倾向于集中在节假日的上午;参与人数达到500人以上、发生在工作日期间的政府单位的群体性突发事件,其主要诱发因素为群体利益诉求的表达。对群体性突发事件特征要素的强关联规则挖掘对进一步分析事件的演化规律和开展精准化的识别预警具有一定实际意义。

In order to solve the problem of low recognition accuracy in early warning and prevention of mass emergencies,by collecting typical domestic mass emergencies in recent years,based on element extraction and structured definitions,the method of association rules analysis of data mining is used to study the relationships among multiple elements such as the time,place,inducing factors,and number of participants.The research results show that there is a strong correlation between the place of occurrence and the date and the specific time.Among them,the Group Emergencies that occur in the government and enterprises mainly occur in the afternoon of the working day;the Group Emergencies that occur in the public places with the main purpose of expressing interest appeals tend to be concentrated in the morning of holidays;the main inducing factor of Group Emergencies in government units with more than 500 participants and occurring on working days is the expression of group interest demands.The mining of strong association rules for the characteristic elements of group emergencies has certain practical significance for further analysis of the evolution of events and the development of accurate identification and early warning.