Show simple item record

dc.rights.licenseCC-BY-NC-ND
dc.contributor.advisorVan der Zee, E.L.
dc.contributor.authorRoozen, B.
dc.date.accessioned2021-07-26T18:00:13Z
dc.date.available2021-07-26T18:00:13Z
dc.date.issued2021
dc.identifier.urihttps://studenttheses.uu.nl/handle/20.500.12932/39867
dc.description.abstractAs cities get bigger and more crowded, questions arise on how to deal with the issues and challenges that come paired with it. Knowledge of the exact situation may help to get an insight into the situation and potentially adjust policies. For this, it is required to gain understanding in the spatial and temporal behavior of individuals. Simply put, an answer is needed to the question “Where are people, and when are they there?” Over the last years, advancements in data collection, storage and analysis have been made. People are posting pictures, sharing information, and reviewing meals when they have been at a certain place. All this data is stored in the social media applications where they were posted to. When the posts are provided with a geotag and a timestamp, this geosocial data could be used to trace the digital footsteps of individuals. Therefore, the aim of this study is to investigate how different sources of geosocial data can be used to visualize both spatial and temporal patterns of individuals. As this study will be a proof-of-concept of using geosocial data, the study area will consist of an urban area (Rotterdam) and a more rural area (Veere). By doing a thorough investigation, the choice is made for using Instagram and TripAdvisor as geosocial data sources. After an extensive data collection and conversion process, the data can be used for analysis. Displaying the descriptive statistics for both the temporal and spatial results reveal many of the behavioral patterns of people. To add to that, the Moran’s I has been calculated to check for spatial autocorrelation and a hot spot analysis has been performed. This came to numerous interesting results, among which an interesting visualization on an invisible boundary around the city center of Rotterdam. There are however some ifs and buts when working with geosocial data. Concluding, it can be stated that where temporal patterns are better visible in rural Veere, the spatial patterns are better visible in urban Rotterdam. The degree of urbanity must therefore be considered when doing research in this area.
dc.description.sponsorshipUtrecht University
dc.format.extent3902580
dc.format.mimetypeapplication/pdf
dc.language.isoen
dc.titleCRAWLING THE WORLD WIDE WEB TO FIND HIDDEN PATTERNS OF PEOPLE
dc.type.contentMaster Thesis
dc.rights.accessrightsOpen Access
dc.subject.keywordsWeb scraping; UGC; Data collection; Instagram; TripAdvisor; Spatial-temporal patterns; Hot spot analysis
dc.subject.courseuuGeographical Information Management and Applications (GIMA)


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record