特大型城市的公交车上普遍安装了自动售票(auto fare collection,AFC)系统和车载GPS导航定位设备,记录了乘客出行与行车轨迹、时间等数据,连续运行的公交车和众多的出行乘客形成了城市公共交通运行和出行大数据。如何高效、准确地从...特大型城市的公交车上普遍安装了自动售票(auto fare collection,AFC)系统和车载GPS导航定位设备,记录了乘客出行与行车轨迹、时间等数据,连续运行的公交车和众多的出行乘客形成了城市公共交通运行和出行大数据。如何高效、准确地从公交大数据中识别公交乘客下车站点,对于提高交通运行效率、科学布局组织交通具有重要的意义和作用。基于深圳市公交车AFC和GPS数据,利用时间匹配和基于带噪声空间密度聚类的方法判别上车站点;在仅有的公交数据基础上,通过分析乘客出行行为,根据乘客多天的出行以及各站点的上车频率,利用乘客高频站点和下游站点吸引权,提出一种推算方法,实现乘客下车站点的推断。算法模型的检验和实例分析表明该方法的有效性。展开更多
Let X1,X2,... be a sequence of independent random variables (r.v.s) belonging to the domain of attraction of a normal or stable law. In this paper, we study moderate deviations for the self-normalized sum n X ∑^n_i...Let X1,X2,... be a sequence of independent random variables (r.v.s) belonging to the domain of attraction of a normal or stable law. In this paper, we study moderate deviations for the self-normalized sum n X ∑^n_i=1Xi/Vm,p ,where Vn,p (∑^n_i=1|Xi|p)^1/p (P 〉 1).Applications to the self-normalized law of the iteratedlogarithm, Studentized increments of partial sums, t-statistic, and weighted sum of independent and identically distributed (i.i.d.) r.v.s are considered.展开更多
文摘特大型城市的公交车上普遍安装了自动售票(auto fare collection,AFC)系统和车载GPS导航定位设备,记录了乘客出行与行车轨迹、时间等数据,连续运行的公交车和众多的出行乘客形成了城市公共交通运行和出行大数据。如何高效、准确地从公交大数据中识别公交乘客下车站点,对于提高交通运行效率、科学布局组织交通具有重要的意义和作用。基于深圳市公交车AFC和GPS数据,利用时间匹配和基于带噪声空间密度聚类的方法判别上车站点;在仅有的公交数据基础上,通过分析乘客出行行为,根据乘客多天的出行以及各站点的上车频率,利用乘客高频站点和下游站点吸引权,提出一种推算方法,实现乘客下车站点的推断。算法模型的检验和实例分析表明该方法的有效性。
基金supported by Hong Kong Research Grant Committee (Grant Nos.HKUST6019/10P and HKUST6019/12P)National Natural Science Foundation of China (Grant Nos. 10871146 and 11271286)the National University of Singapore (Grant No. R-155-000-106-112)
文摘Let X1,X2,... be a sequence of independent random variables (r.v.s) belonging to the domain of attraction of a normal or stable law. In this paper, we study moderate deviations for the self-normalized sum n X ∑^n_i=1Xi/Vm,p ,where Vn,p (∑^n_i=1|Xi|p)^1/p (P 〉 1).Applications to the self-normalized law of the iteratedlogarithm, Studentized increments of partial sums, t-statistic, and weighted sum of independent and identically distributed (i.i.d.) r.v.s are considered.