Repository logo
 

Differentiating population spatial behaviour using a standard feature set

dc.contributor.advisorStanley, Kevin
dc.contributor.committeeMemberGutwin, Carl
dc.contributor.committeeMemberMondal, Debajyoti
dc.contributor.committeeMemberDiab, Ehab
dc.creatorZhang, Rui 1990-
dc.date.accessioned2019-07-23T17:38:45Z
dc.date.available2019-07-23T17:38:45Z
dc.date.created2019-11
dc.date.issued2019-07-23
dc.date.submittedNovember 2019
dc.date.updated2019-07-23T17:38:45Z
dc.description.abstractMoving through space, consuming services at locations, transitioning and dwelling are all aspects of spatial behavior that can be recorded with unprecedented ease and accuracy using the GPS and other sensor systems on commodity smartphones. Collection of GPS data is becoming a standard experimental method for studies ranging from public health interventions to studying the browsing behavior of large non-human mammals. However, the millions of records collected in these studies do not lend themselves to traditional geographic analysis. GPS records need to be reduced to a single feature or combination of features, which express the characteristic of interest. While features for spatial behavior characterization have been proposed in different disciplines, it is not always clear which feature should be appropriate for a specific dataset. The substantial effort on subjective selection or design of feature may or may not lead to an insight into GPS datasets. In this thesis we describe a feature set drawn from three different mathematical heritages: buffer area, convex hull and its variations from activity space, fractal dimension of the recorded GPS traces, and entropy rate of individual paths. We analyze these features against six human mobility datasets. We show that the standard feature set could be used to distinguish disparate human mobility patterns while single feature could not distinguish them alone. The feature set can be efficiently applied to most datasets, subject to the assumptions about data quality inherent in the features.
dc.format.mimetypeapplication/pdf
dc.identifier.urihttp://hdl.handle.net/10388/12204
dc.subjecthuman spatial behaviour, feature set
dc.titleDifferentiating population spatial behaviour using a standard feature set
dc.typeThesis
dc.type.materialtext
thesis.degree.departmentComputer Science
thesis.degree.disciplineComputer Science
thesis.degree.grantorUniversity of Saskatchewan
thesis.degree.levelMasters
thesis.degree.nameMaster of Science (M.Sc.)

Files

Original bundle
Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
ZHANG-THESIS-2019.pdf
Size:
6.62 MB
Format:
Adobe Portable Document Format
License bundle
Now showing 1 - 1 of 1
No Thumbnail Available
Name:
LICENSE.txt
Size:
2.26 KB
Format:
Plain Text
Description: