| EDT / New York | CET / Rome – Berlin | |
|---|---|---|
| 8:30 – 8:40 | 14:30 – 14:40 | OPENING |
| 8:40 – 9:35 | 14:40 – 15:35 | Martin Gerlach (keynote) |
| 9:35-9:55 | 15:35 – 15:55 | Miguel Aguilera |
| 9:55-10:15 | 15:55 – 16:15 | Xabier E. Barandiaran |
| 10:15-10:30 | 16:15 – 16:30 | BREAK |
| 10:30-11:25 | 16:30 – 17:25 | David García (keynote) |
| 11:25-11:45 | 17:25 – 17:45 | Alexandre Bovet |
| 11:45-12:05 | 17:45 – 18:05 | Ignacio Guerrero-Martínez |
| 12:05-12:25 | 18:05 – 18:25 | Milad Ranjbar |
| 12:25-12:45 | 18:25 – 18:45 | FINAL REMARKS |
Speaker (keynote): David García
Title: The politicisation of medical topics on social media
Social media brings both opportunities and risks to our society: it can enable networked action and facilitate information sharing but it can also fuel the spread of misinformation and distort public discussions about important scientific and political issues. I will present two empirical studies of social media data that illustrate how medical topics can become heavily politicized on social media. First, I will present the case of the Twitter backlash to the EAT-Lancet report about sustainable diets, a case that illustrates the presence of new political divides in dietary habits. Second, I will present the analysis of Twitter sharing of medical preprints, which shows a sharp increase in both attention and politicization of the discussion around clinical and research documents since the beginning of the COVID-19 pandemic. This kind of emergence of political divides in new topics can be explained by a computational model based on cognitive science principles that does not require the existence of filter bubbles or echo chambers. Our research shows how partisan voices can dominate the online discussion on health-related topics, highlighting polarized structures and dormant conflicts that difficult science communication through social media. I discuss these results within a new approach to participatory propaganda and misinformation that focuses on understanding how new ontologies of truth can threaten democratic information spaces.
Speaker (keynote): Martin Gerlach
Title: The science of knowledge equity at the Wikimedia Foundation
Today Wikipedia has over 55M articles available across more than 300 different language editions. With roughly 20 billion page views every month, it not only is the main source of information for many users but it has also become a canonical part of the free knowledge ecosystem outside of Wikipedia. As a social movement, the Wikimedia movement has defined knowledge equity as one of its foundational principles towards breaking down the social, political, and technical barriers preventing people from accessing free knowledge. In this talk, I will give an overview of different research directions within the Wikimedia Foundation to support knowledge equity by addressing knowledge gaps and ensuring the integrity of its content. I will describe how using tools from machine learning, network science, and NLP allows to develop new methods for characterizing bias and reliability of content in Wikipedia. I will close discussing our efforts to grow and support a network and community of researchers by developing and sharing new tools and datasets among other things.
Speaker: Alexandre Bovet
Authors: Stuart Feldman, James Flamino, Alessandro Galeazzi, Brendan Cross, Zhenkun Zhou, Matteo Serafino, Alexandre Bovet, Hernán A. Makse, Boleslaw K. Szymanski
Title: Twitter Influencers and Increased Media Polarization during two U.S. Presidential Elections
New social media are decentralized, interactive, and transformative by empowering users to produce and spread information to influence others. In short, they changed the ways information and influence spread. Here, we analyze dynamics of political polarization among Twitter users using hundreds of millions of tweets that we collected over the 2016 and 2020 US presidential elections. From this data, we recreate news diffusing Twitter retweet networks segregated by political orientations. We identify top influencers in each news category in terms of their ability to spread information. The top influencers are classified into those affiliated with a traditional media organization, or with a political organization, or unaffiliated. Most of the top influencers were affiliated with media organizations during both elections. We measure the strength of polarization among them using different methods to assure robustness of the result. We find a clear increase of their polarization from 2016 to 2020. We also find that 75% of the top 100 influencers of all media categories in 2020 were not there in 2016, demonstrating how difficult it is to retain influencer status. The majority of influencers affiliated with traditional media shrunk their fraction by 10% from 2016 to 2020. The replacement came mostly from influencers affiliated with political organizations with center or right orientations. Independent influencers advanced too, overtaking about one third of the majority drop. These and other results create a foundation for understanding how new social media increase polarization and transform the election process.
Speaker: Ignacio Guerrero-Martínez
Authors: Ignacio Guerrero-Martínez, Guillem Suau-Gomila
Title: Covid-19 Communication in Social Media: An Analysis of Media and Institutional Accounts in Facebook and Instagram
The year 2020 has been marked by the Covid-19 pandemic that has caused more than
80,000 deaths and a State of Alarm in Spain from March 2020 to May 2021, the most serious and lasting State of Alarm of the Spanish democracy. In this context, the population has turned to social networks to find out about the Coronavirus. Consequently, the main objective of our study is to analyze the coverage of Covid-19 carried out by media and institutional profiles, as well as the interaction of the public with this information through Facebook and Instagram. We have extracted our analysis sample with the specialized software Crowdtangle, which contains publications made in a period that goes from March 11, 2020 (declaration of Covid-19 as a pandemic) until May 9, 2021 (end of the State of Alarm). Subsequently, we have filtered these data through the sampling system, Top Discussion Indicator (TDI). The study is based on a quantitative analysis based on social media metrics. Our main conclusion is that media profiles have viralized their information more than institutional profiles, that is, they have achieved greater attention from citizens on digital social networks.
Speaker: Miguel Aguilera
Authors: Miguel Aguilera
Title: Rhythms of the Collective Brain: Metastable Synchronization and Cross-Scale Interactions in the Spanish 15M movement
Crowd behaviour challenges our fundamental understanding of social phenomena. Involving complex interactions between multiple temporal and spatial scales of activity, its governing mechanisms defy conventional analysis. Using 1.5 million Twitter messages from the 15M movement in Spain as an example of multitudinous self-organization, we describe the coordination dynamics of the system measuring phase-locking statistics at different frequencies using wavelet transforms, identifying 8 frequency bands of entrained oscillations between 15 geographical nodes. Then, we apply maximum entropy inference methods to describe Ising models capturing transient synchrony in our data at each frequency band. The models suggest that (1) all frequency bands of the system operate near critical points of their parameter space and (2) while fast frequencies present only a few metastable states displaying all-or-none synchronization, slow frequencies present a diversity of metastable states of partial synchronization. Furthermore, describing the state at each frequency band using the energy of the corresponding Ising model, we compute transfer entropy to characterize cross-scale interactions between frequency bands, showing (1) a cascade of upward information flows in which each frequency band influences its contiguous slower bands and (2) downward information flows where slow frequencies modulate distant fast frequencies.
Full paper: https://www.hindawi.com/journals/complexity/2018/4212509/
Speaker: Xabier Barandiaran
Authors: Xabier E. Barandiaran, Antonio Calleja-López and Emanuele Cozzo
Title: An operational definition of collective identities in digital interaction networks. Theory and applications
We are currently witnessing the emergence of new forms of collective identities and a redefinition of the old ones through networked digital interactions, and these can be explicitly measured and analyzed. We distinguish between three major trends on the development of the concept of identity in the social realm: (1) an essentialist sense (based on conditions and properties shared by members of a group), (2) a representational or ideational sense (based on the application of categories by oneself or others), and (3) a relational and interactional sense (based on interaction processes between actors and their environments). The interactional approach aligns with current empirical and methodological progress in social network analysis. Moreover, it has been argued that, within the network society, the notion of collective identity in the political field must be rethought as technologically mediated and interactive. We suggest that collective identities should be understood as recurrent, cohesive, and coordinated communicative interaction networks. We here propose that such identities can be depicted by: (a) mapping and filtering a relevant interaction network, (b) delimiting a set of communities, (c) determining the strongly connected component(s) of such communities (the core identity) in a directed graph, and (d) defining the identity audiences and sources within the community. This technical graph–theoretical characterization is explained and justified in detail through a toy model and applied to three empirical case studies to characterize political identities in party politics (communicative interaction in Twitter during the Spanish elections in 2018), contentious politics in confrontation (in Twitter during the Catalan strike for independence 2019), and the multitudinous identity of Spanish Indignados/15M social movement (in Facebook fan pages 2011). We discuss how the proposed definition is useful to delimit and characterize the internal structure of collective identities in technopolitical interaction networks, and we suggest how the proposed methods can be improved and complemented with other approaches. We finally draw the theoretical implications of understanding collective identities as emerging from interaction networks in a progressive platformization of social interactions in a digital world.
Full paper: https://www.frontiersin.org/articles/10.3389/fpsyg.2020.01549/full
Speaker: Milad Ranjbar
Authors: Milad Ranjbar, Fakhteh Ghanbarnejad, Sadegh Raeisi
Title: Retweet Dynamics
Social media has become one of the most important mediums for communication in recent years, and studying the dynamics of information on these platforms has attracted a lot of attention. In this paper we focus on retweeting dynamics. Retweeting could be considered as propagation of a news, an opinion, etc.. We investigate the role of individuals on the dynamics of retweets. To this end, we build a machine learning model. The model takes the attributes of two individuals on the network and predicts the probability of the second one (target) retweeting the message from the first one (source). We refer to the two possible outcomes as active or inactive edges. We train machine learning models on tweets from three different languages, English, German, and Persian(Farsi). These languages could represent different communities with different cultures. In all these languages we can predict retweeted edges with at least 80% accuracy. We also investigate the most effective attributes which shape this retweeting behaviour. We find out that the number of the followers of the tail agent and how active the head agent is, have the most impact on activating the edge between them. According to our analysis we conclude that retweeting behaviour could be more influenced by individuals’ behaviour rather than the content of the tweets. These findings provide a different perspective on human behaviour on social media.
PCNet 