Dartboard-Shaped Spatial Segmentation: Depicting Distribution of Geo-Tweets and Defining Urban Boundaries  
Author Jaehee Cho


Co-Author(s) Euiyoung Baik


Abstract This study compares and analyzes the spatial distribution of people in three Vietnamese cities using the geolocated Twitter data. Ho Chi Minh, Ha Noi and Da Nang were selected as the target cities. The data collection period is 123 days in 2016 and 125 days in 2018. The spatial distribution of the three cities is analyzed according to two periods and residence status (local vs foreigner). In this study, we have found popular places based on a newly devised spatial segmentation method and we also estimated the urban boundaries by reflecting the location as well as the volume of geo-tweets. Instead of histograms, CDF (Cumulative Distribution Function) curves are employed to show the distribution of distances between the city center point and actual tweet location.


Keywords Geo-located tweet, Spatial segmentation, Urban boundary, Cumulative distribution function
    Article #:  DSBFI19-74
Proceedings of ISSAT International Conference on Data Science in Business, Finance and Industry
July 3-5, 2019 - Da Nang, Vietnam