Professional drivers are more frequently exposed to longer driving distance and travel time,leading to a higher possibility of safety risk for distraction and fatigue.The widespread and common use of commercial driver...Professional drivers are more frequently exposed to longer driving distance and travel time,leading to a higher possibility of safety risk for distraction and fatigue.The widespread and common use of commercial driver monitoring systems(DMS)provides a potential for data collection.It increases the amount of data characterizing driver behavior that can be used for further safety research.This study utilized DMS warning-based data and applied an association rule mining approach to explore risk factors contributing to hazardous materials(HAZMAT)truck driver inattention.A total of 499 HAZMAT truck driver inattentive warning events were used to find rules that will predict the occurrence of driver’s fatigue and distraction.First,Fisher’s exact tests were performed to examine the association between the frequency of driver inattentive behavior warnings and risk factors.Second,support,confidence,and lift values were used as measurements to quantify the relative strength of the association rules generated by the Apriori algorithm.Results show that speed between 40and 49 km/h,relatively longer travel time(3-6 h),freeway,tangent section,off-peak hour and clear weather condition are found to be highly associated with fatigue driving,while nighttime during 18:00 to 23:59,speed between 70 and 80 km/h,travel time between 1 and 3 h,freeways,acceleration less than 0.5 m/s^(2),visibility greater than 1000 m,and tangent roadway section are found to be highly associated with distracted driving.By focusing on the specific feature groups,these association rules would help in the development of mitigating distraction and fatigue driving countermeasures and enforcement approaches.展开更多
The problem of association rule mining has gained considerableprominence in the data mining community for its use as an important tool of knowledge discovery from large-scale databases. And there has been a spurt of r...The problem of association rule mining has gained considerableprominence in the data mining community for its use as an important tool of knowledge discovery from large-scale databases. And there has been a spurt of researchactivities around this problem. Traditional association rule mining is limited tointratransaction. Only recently the concept of N-dimensional inter-transaction association rule (NDITAR) was proposed by H.J. Lu. This paper modifies and extendsLu's definition of NDITAR based on the analysis of its limitations, and the generalized multidimensional association rule (GMDAR) is subsequently introduced, whichis more general, flexible and reasonable than NDITAR.展开更多
基金supported by National Key R&D Program of China(2021YFC3001500).
文摘Professional drivers are more frequently exposed to longer driving distance and travel time,leading to a higher possibility of safety risk for distraction and fatigue.The widespread and common use of commercial driver monitoring systems(DMS)provides a potential for data collection.It increases the amount of data characterizing driver behavior that can be used for further safety research.This study utilized DMS warning-based data and applied an association rule mining approach to explore risk factors contributing to hazardous materials(HAZMAT)truck driver inattention.A total of 499 HAZMAT truck driver inattentive warning events were used to find rules that will predict the occurrence of driver’s fatigue and distraction.First,Fisher’s exact tests were performed to examine the association between the frequency of driver inattentive behavior warnings and risk factors.Second,support,confidence,and lift values were used as measurements to quantify the relative strength of the association rules generated by the Apriori algorithm.Results show that speed between 40and 49 km/h,relatively longer travel time(3-6 h),freeway,tangent section,off-peak hour and clear weather condition are found to be highly associated with fatigue driving,while nighttime during 18:00 to 23:59,speed between 70 and 80 km/h,travel time between 1 and 3 h,freeways,acceleration less than 0.5 m/s^(2),visibility greater than 1000 m,and tangent roadway section are found to be highly associated with distracted driving.By focusing on the specific feature groups,these association rules would help in the development of mitigating distraction and fatigue driving countermeasures and enforcement approaches.
文摘The problem of association rule mining has gained considerableprominence in the data mining community for its use as an important tool of knowledge discovery from large-scale databases. And there has been a spurt of researchactivities around this problem. Traditional association rule mining is limited tointratransaction. Only recently the concept of N-dimensional inter-transaction association rule (NDITAR) was proposed by H.J. Lu. This paper modifies and extendsLu's definition of NDITAR based on the analysis of its limitations, and the generalized multidimensional association rule (GMDAR) is subsequently introduced, whichis more general, flexible and reasonable than NDITAR.
