Tracking the political opinion landscape in Twitter during electoral periods

MUSSI REYERO, TOMÁS; MARIANO G. BEIRO; ALVAREZ-HAMELIN, J. IGNACIO; HERNÁNDEZ, LAURA; KOTZINOS, DIMITRIS

Lyon

Conferencia; CONFERENCE ON COMPLEX SYSTEMS 2021; 2021

COMPLEX SYSTEMS SOCIETY

An interesting way to study the structure of opinions in Twitter exploits the hashtags chosenby the users, assuming that this choice reveals a concept that the user wishes to address. In arecent work [1], topics are defined by determining the community structure in a weightednetwork of hashtags, where two hashtags are connected if they appear together in the sametweet. Assuming that the coexistence of hashtags is semantically meaningful, the communitystructure of such network can reveal the general topics under discussion. In this way, the usersmay be characterized by a topic vector, with a dimension equal to the number of communitiesdetected and where each component informs about whether the user is interested more or lessthan the average over the population, on the different topics.In this work we extend these ideas to a dynamical study of the rapidly evolving opinionlandscape that takes place in a society during an electoral campaign and with the case study ofthe two last presidential elections in Argentina (2015 and 2019). Specifically we understandby evolution of opinion, the evolution of the preferences of the social actors concerning thetopics discussed in the platform, and we are particularly interested in detecting whetherspecific groups or users synchronize around some specific topics at some given times.We compute similarity metrics among supporters of a given political party and acrossmembers of different parties as a function of time, and we are able to detect in a completelyagnostic way (without choosing topics to follow) if the groups synchronize at a given time,meaning that they are all talking about the same topics, or on the contrary when they enter incomplete opposition.Our results show that the method is able to capture the reshaping of the political opinionlandscape which has led to the inversion of the results between the two rounds of 2015Argentinian election.