Exploring human mobility patterns based on geotagged Flickr photos

Thumbnail Image
Files
master_Raja_Muneeba_2015.pdf (6.48 MB)
Journal Title
Journal ISSN
Volume Title
Insinööritieteiden korkeakoulu | Master's thesis
Unless otherwise stated, all rights belong to the author. You may download, display and print this publication for Your own personal use. Commercial use is prohibited.
Author
Date
2015-11-23
Department
Major/Subject
Geoinformation Technology
Mcode
IA3002
Degree programme
Geomatiikan koulutusohjelma
Language
en
Pages
v+62
Series
Abstract
Predicting human mobility behaviour has long been a topic of scientific interest. Such studies generally rely on tracking human movements through a range of data collection methodologies such as using GPS trackers, cellular network data etc. Some of this data may be confidential or hard to acquire. This thesis explores if existing publicly available data on online photo sharing platforms can be used to determine human mobility patterns with reasonable accuracy. We choose the Flickr website as the data collection medium as it has an extensive user base actively sharing photos many of which, have geo tags embedded in them which are preserved by Flickr. Our analysis reveals that while the data from Flickr is sparse and discontinuous making it unsuitable for reliable mobility prediction, typical human mobility trends based on time of day, day of week and month of the year can still be extracted. Such interesting patterns could be potentially used in traffic engineering domains or for user profiling purposes. More specifically, we describe how to obtain a subset of frequent active users and their information from Flickr, and the sliding window mechanism to filter the active periods of the users. Later we explain the various statistical methods applied on the filtered subset of data to identify the categories in which users could be classified, mainly short distance travellers and long distance travellers. The short distance travellers are considered for mobility trends prediction.
Description
Supervisor
Virrantaus, Kirsi
Thesis advisor
Ott, Jorg
Hyytia, Esa
Keywords
pattern mining, movement patterns, micro-mobility, spatial-temporal analysis, Flickr geotagged images
Other note
Citation
Permanent link to this item
https://urn.fi/URN:NBN:fi:aalto-201512165581
Collections
[dipl] Insinööritieteiden korkeakoulu / ENG