KIT | KIT-Bibliothek | Impressum | Datenschutz
Originalveröffentlichung
DOI: 10.1007/978-3-319-29009-6_4

On Spatial Measures of Geographic Relevance for Geotagged Social Media Content

Wang, Xin; Gaugel, Tristan; Keller, Matthias

Abstract:
Recently, geotagged social media contents became increasingly available to researchers and were subject to more and more studies. Different spatial measures such as Focus, Entropy and Spread have been applied to describe geospatial characteristics of social media contents. In this paper, we draw the attention to the fact that these popular measures do not necessarily show the geographic relevance or dependence of social content, but mix up geographic relevance, the distribution of the user population, and sample size. Therefore, results based on these measures cannot be interpreted as geographic effects alone. By means of an assessment, based on Twitter data collected over a time span of six weeks, we highlight potential misinterpretations and we furthermore propose normalized measures which show less dependency on the underlying user population and are able to mitigate the effect of outliers.


Zugehörige Institution(en) am KIT Institut für Telematik (TM)
Steinbuch Centre for Computing (SCC)
Publikationstyp Proceedingsbeitrag
Jahr 2016
Sprache Englisch
Identifikator ISBN: 978-3-319-29008-9
ISSN: 0302-9743
KITopen-ID: 1000052804
Erschienen in Big Data Analytics in the Social and Ubiquitous Context. 5th International Workshop on Modeling Social Media, MSM 2014, 5th International Workshop on Mining Ubiquitous and Social Environments, MUSE 2014, and First International Workshop on Machine Learning for Urban Sensor Data, SenseML 2014, Revised Selected Papers. Ed.: F. Janssen
Verlag Springer, Cham
Seiten 70-89
Serie Lecture Notes in Computer Science ; 9546
Nachgewiesen in Scopus
KIT – Die Forschungsuniversität in der Helmholtz-Gemeinschaft KITopen Landing Page