从智能卡数据挖掘客流信息的算法  

An Algorithm for Mining Passenger Flow Information from Smart Card Data

作　　者：高联雄[1,2] 吴建平[1] 

机构地区：[1]北京邮电大学计算机学院,北京100876 [2]云南民族大学电气信息工程学院,昆明650093

出　　处：《北京邮电大学学报》2011年第3期94-97,共4页Journal of Beijing University of Posts and Telecommunications

摘　　要：为了从广泛使用的智能卡付费系统获取公交客流信息,提出了一种利用公交调度信息和智能卡刷卡信息推断乘客上车站点的方法.对同一辆车的连续2次刷卡进行朴素贝叶斯分类,区分是否是在同一个站刷卡;利用极大似然估计、动态规划和二次规划方法估计出各路段的行程时间;运用坐标下降法从不准确的初始参数出发,交替估计行程时间和行程时间的参数,从而推断出每次刷卡的上车站点.实验结果验证了新方法的正确性和有效性,证明了该方法误差较小,收敛较快.To collect passenger flow information of public transit from the widely applied payment systems, a new method is proposed to infer the stops at which passengers hold board the bus from smart card fare data and bus schedules. The method first classifies smart card fare ing smart cards two sequential swipes to decide whether they occur at the same stop with naive Bayes classifier. Travel times are then estimated from the naive Bayes classifier results using maximum likelihood estimation, dynamic program- ming and quadratic programming methods. To solve the problem with imprecise initial parameters, a coordinate descent method is applied. It updates parameters and estimates values alternatively until convergence. An experiment is designed to test this algorithm with real-world data, and it proves that the error of this method is small and the convergence is fast.

关 键 词：公共交通网络 公交智能卡 朴素贝叶斯分类 动态规划 二次规划 坐标下降法 

分 类 号：U491.17[交通运输工程—交通运输规划与管理]