Department of Geography - University of Zurich
Conducted by Linus Rüegg and Eric Tharmlingam Mentored by Carlota de Benito Moreno and Peter Ranacher
In this project, we close geographic data-gaps and visualize the spatio-temporal spread of linguistic innovations in a social and geographic network. Linguistic innovations describe new phenomena in languages, such as changes or creation of expressions, their notion, or their spelling. To create our networks, we track the spread of 33 innovations in the Spanish-speaking Twitter community.
The geolocation accuracy of the tweets and their users vary from exact coordinates, over city names, to no geographic information at all. Therefore, only very coarse visualizations of the geographic network in the dataset were possible. Our main objectives are to 1) create comprehensive visualizations of the dataset, 2) and to offer a method to close geographic gaps in the spatio-temporal network.
To contrive our objectives, we use a dataset containing all tweets in Spanish of the first week of every month from 2012 to 2019.