摘要
针对当前网约车服务质量提升需求,本文研究的重点是如何更好的发挥网约车在城市交通的作用和效果。对此,本文以陕西西安某网约车公司的GPS数据作为研究基础,采用K均值聚类的方法完成对居民用户出行的时空特征分析,然后采用等待时间推荐模型完成对某区域内空网约车的推荐。最后,通过试验对上述的方案进行验证,结果表明在研究区域内,西安市民工作日出行的密度明显高于非工作日的出行密度,同时通过本文构建的乘客推荐算法,乘客平均等待的时间要明显低于传统的推荐算法。由此说明通过时空挖掘可很好的统计居民出行规律,同时通过推荐算法可大大减少用户等待时间,为市民出行提供方便。
In view of the current demand for improving the service quality of network-based cars,this paper focuses how to better play the role and effect of network-based cars in urban traffic.In this regard,based on the GPS data of a company in Xi'an, this paper uses the method of K means clustering to analyze the spatial and temporal characteristics of the travel of the residents, and then uses the waiting time recommendation model to complete the recommendation of the empty network vehicles in a certain area.Finally,the above scheme is verified by experiments.The results show that the travel density of Xi'an citizen working day is obviously higher than that of non-working day.At the same time,the average waiting time of passengers is obviously lower than the traditional recommendation algorithm through the passenger recommendation algorithm constructed in this paper.It also shows that the travel time of passengers is obviously lower than that of the traditional recommendation algorithm. Spatio-temporal mining can be a good statistic of residents'travel rules,while recommendation algorithm can greatly reduce the waiting time for users,providing convenience for citizens to travel.
作者
贾步忠
JIA Buzhong(College of AccountingⅡ,Shanxi Technical College of Finance &Economics,Xianyang 712000)
出处
《微型电脑应用》
2018年第12期83-86,共4页
Microcomputer Applications
关键词
网约车
GPS数据
K均值聚类
时空特征
推荐模型
Network-baed cars
GPS data
K means clustering
Spatial and temporal characteristics
Recommendation model
