移动端个体出行链数据自采设计及出行特征选择

Data Mining of Individual Travel Chain Based on Mobile Phone and Travel Characteristics

手机信令数据和GPS数据等移动互联网位置数据的应用,为获取完整的个体出行链提供了可能。基于移动互联网位置数据提取的出行特征及建立的出行分析模型,与真实的出行情况是否匹配,需要被测试。因此,有必要通过移动互联网位置数据采集和交通调查两种方式,获取同一个体完整的出行链数据,用交通调查数据来标定出行分析结果。设计一种基于移动端的个体出行链数据自采方案,同时采集个体的GPS数据、基站位置数据以及出行链记录数据。在出行特征分析的基础上,应用主成分分析和决策树两种方法,筛选可用于交通方式识别的关键特征。结果表明,速度最大值、速度均值、加速度标准差、速度标准差和方位角变化标准差可显著区分交通方式。

The application of mobile internet location data,such as cell phone GSM data and GPS data,makes it possible to obtain a complete individual travel chain.Whether the travel characteristics extracted from mobile internet location data and the established travel analysis model match the realworld situation needs to be tested.Therefore,it is necessary to obtain the complete travel chain data of the same individual through two ways of mobile internet location data collection and traditional travel survey,and use travel survey data to calibrate travel analysis results.This paper designs an individual travel chain data acquisition scheme based on cell phone,and collects three types of data simultaneously,including individual GPS data,base station location data and travel chain.Based on the analysis of travel characteristics,two methods of principal Component Analysis(PCA)and decision tree-based methods are applied to screen key characteristics that can be used for travel mode identification.The results show that the maximum speed,average speed,standard deviation of acceleration,standard deviation of speed and standard deviation of azimuth change can distinguish travel modes significantly.