Please use this identifier to cite or link to this item:
Twitter Connections Shaping New York City
|Title:||Twitter Connections Shaping New York City|
show 1 moreKontokosta, Constantine
|Keywords:||Deep Learning, Ubiquitous and Toy Computing|
Human mobility, social networks, big data, social media, community detection
|Issue Date:||03 Jan 2018|
|Abstract:||Geo-tagged Twitter has been proven to be a useful proxy for urban mobility, this way helping to understand the structure of the city and the shape of its local neighborhoods. In the present work we approach this problem from another angle by leveraging additional information on Twitter customers mentioning each other, which might partially reveal their social relations. We propose a novel way of constructing a spatial social network based on such data, analyze its structure and evaluate its utility for delineating urban neighborhoods. This delineation happens to have substantial similarity to the earlier one based on the user mobility network. It leads to an assumption that the social connectivity between the users is strongly related with the similarity in their mobility patterns. We justify this hypothesis enabling extrapolation of the available user mobility patterns as a proxy for social connectivity and building a network of hidden ties based on the mobility pattern similarity. Finally, we evaluate the socio-economic characteristics of the partitions for all three networks of all mentioning, reciprocal mentioning and the hidden ties.|
|Rights:||Attribution-NonCommercial-NoDerivatives 4.0 International|
|Appears in Collections:||Deep Learning, Ubiquitous and Toy Computing|
Please contact firstname.lastname@example.org if you need this content in an alternative format.
Items in ScholarSpace are protected by copyright, with all rights reserved, unless otherwise indicated.