Reproductibilitat de les anàlisis de xarxes de Twitter
Year: 2022
Type: dissertation
Abstract: Twitter is a popular microblogging social network that allows researchers to easily download datasets of tweets to analyze them. This has made it the data source of a large number of studies and, in particular, analyses of interaction networks between users. One of the ways to download Twitter data it is with the Standard Search API. But there is the limitation that only allows to download tweets that they are less than a week old. This seriously compromises the possibility of reproducing research done with this data. Recently, however, Twitter has released a new API for academic research. With this new Academic Search API, tweets written at any time can be downloaded. In this work, we use the new API to repeat queries done years ago. So, for every query, we have two sets of tweets, those that were originally downloaded and those that we downloaded this year. For every set of tweets, we build a user graph based on their retweets. To build it we use Twarc Network, a tool we have helped to develop during the realization of this work. We compare every pair of sets of tweets and graphs to see which properties changed the most.
Author: Massachs Güell, Joan
Language: English
Subfield: Statistical and Nonlinear Physics
Field: Physics and Astronomy
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W7111873425