基于出租车GPS轨迹数据挖掘的居民出行特征研究

龙源期刊网 https://www.wendangku.net/doc/e0973241.html,

基于出租车GPS轨迹数据挖掘的居民出行特征研究

作者：林基艳张雅琼张慧

来源：《计算机时代》2017年第05期

摘要：城市居民出行特征分析在城市交通规划中日益重要，已经成为城市道路交通建设的一个重要依据，而城市出租车因为其运营特点和规律，可以很好地反应出居民出行的特征。因此，为了有效地解决城市道路拥堵问题、为城市交通规划提供依据，论文提出了基于出租车GPS轨迹数据挖掘的居民出行特征研究，该研究利用DBSCAN算法对出租车历史GPS运行轨迹数据进行聚类分析，不仅可以提取城市居民出行的时空特征，也可以有效地降低出租车的空载率。

关键词：居民出行特征；数据挖掘； GPS轨迹数据； DBSCAN

中图分类号：TP29 文献标志码：A 文章编号：1006-8228（2017）05-37-03

Research on the characteristics of resident travel based on the taxi

GPS trajectory data mining

Lin Jiyan， Zhang Yaqiong， Zhang Hui

（School of Information Technology， Yulin University， Yulin， Shaanxi 719000， China）

Abstract： The analysis to the characteristics of urban residents travel is becoming increasingly important in urban traffic planning， and has become an important basis of urban road traffic construction. The urban taxi can well reflect the characteristics of residents travel because of its operational characteristics and rule. Therefore， in order to effectively solve the problem of city road congestion and provide the basis for city traffic planning， the paper presents the research on the characteristics of resident travel based on the taxi GPS trajectory data mining. The research uses DBSCAN algorithm to realize the clustering analysis of the historical GPS trajectory data， which can not only extract the temporal and spatial characteristics of urban resident travel， but also effectively reduce the taxi no-load rate.

Key words： resident travel characteristic； data mining； GPS trajectory data； DBSCAN

0 引言