Joint and individual variations in heterogeneous traffic data sets
With a rise in the population of the world’s cities, understanding the dynamics of the commuters’ transportation patterns has become crucial in planning and management of urban facilities and services. In this paper, we explore two novel data mining techniques, namely, Joint and Individual Variation...
Saved in:
Main Author: | |
---|---|
Other Authors: | |
Format: | Final Year Project |
Language: | English |
Published: |
2014
|
Subjects: | |
Online Access: | http://hdl.handle.net/10356/61471 |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Institution: | Nanyang Technological University |
Language: | English |
Summary: | With a rise in the population of the world’s cities, understanding the dynamics of the commuters’ transportation patterns has become crucial in planning and management of urban facilities and services. In this paper, we explore two novel data mining techniques, namely, Joint and Individual Variation Explained (JIVE) for Integrated Analysis of Multiple Data Types, and Common Orthogonal Basis Extraction (COBE), and apply them to smart card data available for passengers in Singapore. Using Origin-Destination (O-D) passenger count matrices for two days of the week for MRT stations in Singapore, we are able to arrive at the “joint” travel patterns, i.e., patterns seen on both days, as well as “individual” travel patterns, i.e., patterns seen only on a particular day but not the other, simply from the O-D pair data for two days. The methodologies to identify, quantify and visualize the travel patterns that have been utilized and their scope been expanded in this project, could aid Transportation Authorities in short-term resource allocation and long-term facilities planning. |
---|