习惯性违章行为耦合关联分析模型

发布时间：2018-02-13 13:59

本文关键词： 习惯性违章行为关联规则(ARM)挖掘思想耦合关系关系特征分析耦合关联分析模型　出处：《中国安全科学学报》2017年02期 　论文类型：期刊论文

【摘要】：为探讨习惯性违章行为(HVB)的属性间与属性内的关系特征,建立习惯性违章行为的耦合关联分析模型。首先,分析违章行为属性值的分布特征及关联关系,运用关联规则(ARM)挖掘思想和耦合关系理论对各类违章行为下相应属性的关联系数进行求解,得到一个耦合关联度向量集,且对诸向量从大到小排序;然后,依据排序后耦合关联度向量集映射成习惯性违章行为耦合关联分析模型。最后,引入召回率、精确率和平均绝对值误差(MAE)等3个指标,分别求解数据集和模型的指标结果。所建模型与ARM分析结果的对比表明,模型在习惯性违章行为关联关系分析的准确性与全面性方面都效果良好。
[Abstract]:In order to study the relationship between the attributes and the attributes of the habitual violation behavior (HVB), a coupling correlation analysis model of the habitual violation behavior is established. Firstly, the distribution characteristics and the correlation relationship of the attribute value of the violation behavior are analyzed. By using association rule mining and coupling relation theory, the correlation coefficients of the corresponding attributes under various illegal behaviors are solved, and a set of coupling correlation degree vectors is obtained, and the vectors are sorted from large to small. According to the ordered coupling correlation degree vector set, the coupled association analysis model of habitual violation behavior is mapped. Finally, three indexes, recall rate, precision rate and mean absolute value error are introduced, such as recall rate, precision rate and average absolute value error. The comparison between the established model and the ARM analysis results shows that the model is effective in the accuracy and comprehensiveness of the relationship analysis of habitual violations.
【作者单位】：辽宁工程技术大学电子与信息工程学院;辽宁省数字化矿山装备工程技术研究中心;辽宁工程技术大学工商管理学院;
【基金】：国家自然科学基金青年项目资助(51504126,51404125) 辽宁省教育厅项目资助(LJYR007)
【分类号】：X91

【相似文献】