Presentation Title

Data Analysis in GPS-Trajectories for Finding the Role of Land Use in Urban Traffic Condition

Faculty Mentor

Kohei Yokoi, Emelinda Parentela

Start Date

18-11-2017 11:00 AM

End Date

18-11-2017 11:15 AM

Location

9-245

Session

Engineering/CS 2

Type of Presentation

Oral Talk

Subject Area

engineering_computer_science

Abstract

Nowadays traffic forecast plays a significant role in improving people’s quality of life at commuting and getting around in a city. Recently, focusing on the relationship between the way of land use and traffic conditions has been increasing its popularity. Given this, we have performed an analysis and visualization of the relationship in the city of Beijing with a dataset that contains GPS-trajectory data collected with taxicabs in the city, and the definition of the road network there.

We have implemented a polynomial regression that is based on the density of each specialized facility; offices, banks, schools and so on, in total ten types to estimate the collected traffic data.

After that, we gained 10 sets of time series of the coefficient index which represented how strongly each facility type became people’s destination at a certain time period. Consequently, two of the series gave us the correlation index between the two facility types, which successfully explained features of specialized facilities we notice in our daily life to some extent.

In conclusion, we verified that focusing on the way of land use in a certain city can examine how the traffic there is composed through the land use. However, we also proved that our data analysis implies there is more efficient model selection for the problem.

Summary of research results to be presented

We have composed a thesis and poster that summarizes this research, and have privately submitted to Keio University, Japan in February 2017, but this application for SCCUR 2017 is permitted by the institution.

The paper and poster are available via the links below.

Paper: https://1drv.ms/b/s!AhcteHcVwUAbjYkYvQx0MT6KqsdKtg

Poster: https://1drv.ms/b/s!AhcteHcVwUAbi9Qg2zTpqnrTdfr4Yg

This document is currently not available here.

Share

COinS
 
Nov 18th, 11:00 AM Nov 18th, 11:15 AM

Data Analysis in GPS-Trajectories for Finding the Role of Land Use in Urban Traffic Condition

9-245

Nowadays traffic forecast plays a significant role in improving people’s quality of life at commuting and getting around in a city. Recently, focusing on the relationship between the way of land use and traffic conditions has been increasing its popularity. Given this, we have performed an analysis and visualization of the relationship in the city of Beijing with a dataset that contains GPS-trajectory data collected with taxicabs in the city, and the definition of the road network there.

We have implemented a polynomial regression that is based on the density of each specialized facility; offices, banks, schools and so on, in total ten types to estimate the collected traffic data.

After that, we gained 10 sets of time series of the coefficient index which represented how strongly each facility type became people’s destination at a certain time period. Consequently, two of the series gave us the correlation index between the two facility types, which successfully explained features of specialized facilities we notice in our daily life to some extent.

In conclusion, we verified that focusing on the way of land use in a certain city can examine how the traffic there is composed through the land use. However, we also proved that our data analysis implies there is more efficient model selection for the problem.