Cross-Platform Digital Networks

Project title: Cross-platform digital networks: Exploring the narrative affordances of force-directed layouts and data relations nature
Lead: Janna Joceli Omena
Team members (alphabetical): Alexandra Deem, Beatrice Gobbo, Daniela Van Geenen, Débora Alves, Elisa Kannasto, Fernanda Benquerer, Germán Llorca-Abad, Géraldine Bengsch, Giacomo Flaim, Herbert Natta, Jason Chao, João Carlos Martins, Lorena Cano-Orón, Marc Smith (NodeXL team), Marcia Lisboa, Paulina Sierra, Saide Mobayed, Verónica Israel Turim.

Project pitch slides 
Final presentation slides

Key Findings
Introduction
Research Questions
Subprojects and Research Questions
Query Design & Visual Protocol
Findings
Discussion
References

Key Findings
Since this project is divided in three cases, the key findings for each case study is presented in the subprojects’ respective pages.

Introduction
The study of digital networks has been occupying a position of prominence in digital humanities and digital social practices.  Current studies have tried to approach digital networks through mixed methods and as an attempt to get the advantage of online data. The application of statistical techniques or sociometric, for instance, has become a common routine to study Internet platforms. However, in such research practices, little attention has been paid to the relational nature of online data, as well as to the platform mechanisms in which online activity is inscribed. As a response to that, this project proposes an alternative way for studying networks, that is questioning how the online activity is inscribed into platforms. Consequently, how to proceed/conduct the process of interpretation and analysis demanded after data collection.

By accounting these matters and considering that different platform data and mechanisms require different ways of reading networks, this project aims to develop the cross-platform digital networks approach (Omena & Amaral, 2019). This approach takes into account a critical framework for reading networks which simultaneously reflects technical-practical knowledge on platform grammatisation, the narrative affordances of ForceAtlas2 and Gephi software. In other words, there is a need for first understanding how platforms delineate and organise online activity, then, the need of grasping the logic of network spatialisation and, finally, how to get an advantage of visualisation software (such as Gephi) to reading the semantic space of networks. 

Types of digital networks

Considering that digital networks can be afforded by technical interfaces (APIS) or online data (Omena & Amaral, 2019), this project conducted research on three types of networks :

  1. Hashtag networks (the hashtag engagement of Twitter, Instagram and Reddit)
  2. Networks of recommendation  (YouTube recommender algorithm)
  3. Computer vision API-based networks (Google Vision API & Clarifai)

Research Questions
With basis on cross-platform data and visual network analysis:

  • How can we explore the affordances of the narrative of the force-directed layouts?

With basis on the cross-platform digital networks approach:

  • What counts when reading different types of networks and what stories their multiple trajectories of visual narratives can tell?
  • Can we guide network reading/interpretation through fixed layers of interpretation (centre, mid-term, periphery and isolated items) and multiple forms of reading?

Subprojects and Research Questions

Project I | #Coronavirus: Social media and the early stages of a global pandemic

How is coronavirus used in different platforms, such as Instagram, Youtube and Facebook?

Project II | Climate change: The networks of #climatechange hashtag engagement & “climate emergency” image circulation

Which are the main hashtags related with #ClimateChange? Co-hashtag
Where are the most retweeted users from?
How is the relation between hashtags? Crossplatform
What can we learn from computer vision APIs-based networks to study the history of “climate emergency” image circulation?

Project III | Masculinity

How is masculinity being represented through different digital platforms use-cultures?

Query Design & Visual Protocol

For each subproject we adopted different strategies and analytical approaches, as well as varied types of networks. For instance, networks of hashtags (e.g. #masculinity, #coronavirus and #climatechange), networks of recommendation (YouTube recommender algorithm about coronavirus, manhood and masculinity, computer vision API-based networks (e.g. Google Image Search results about climate emergency). Google Image Search was used as a pathway for building a computer vision API-based network to study the image circulation related to “climate emergency” over years. Further description and details on the query design can be found in each subproject page.

Findings on reading cross-platform digital networks

The narrative affordances of networks spatialised by ForceAtlas2.

 

This project followed a framework for reading cross-platform digital networks (Omena & Amaral, 2019) which, among other cases, have its basis on previous studies such as reading Google App Store and YouTube recommendation systems, hashtags networks and bot image circulation (see Blaiotta et al., 2018; Christensen et al. 2018; Chao, Pilipets et al. 2019, Omena, 2019). In doing so,  it helped in strengthening the proposal of fixed layers of interpretation (centre, mid-term, periphery and isolated elements) provided by the layout algorithm ForceAtlas2. Regardless of the type of the network, the centre reunites the most connected nodes in relation to the whole network (and in terms of variety and diversity). The mid-term usually gathers the gridging and empty zones, while in the periphery we see particularities (many times these areas proved to be even more interesting than the centre of the network). The isolated nodes, usually left behind in the analytical process, but they exist for a reason, and need attention.

Through this comprehension, we found out that the richness of cross-platform digital networks relies precisely in the understanding of what we read in different networks, that emerge from different platforms or forms of being built up. However, to achieve this level of understanding, the questions of what precedes networks and the relational nature of data and platform mechanism entangled within network visualization should come first. Following this logic, we present some basic guidelines that translate what to read in different types of networks (see the visualization) and, furthermore, the typical visuality of these networks.

Discussion
The proposed technique of reading  cross-platform digital networks through fixed layers of interpretation (centre, mid-term, periphery and isolated nodes) can be a valuable asset that helps the analytical process and work with digital methods. However, before the appropriation of this technique, there are essential prerequisites for reading networks, such as the understanding of different types of networks and how connections are made; the perception of what can exactly be read in different networks, and, of course, minimum technical skills and knowledge. There is, furthermore, a requirement that relates with digital platforms or how online activity is rendered and made available.

References

Blaiotta, E., Hoang, Q. T., Jongeling, M., Omena, J. J., Terenghi,  G., Zu, W. (2018). Mapping war atrocities across platforms. DMI Summer School 2018. Retrieved from  https://wiki.digitalmethods.net/Dmi/SummerSchool2018MappingWarAtrocities

Christensen, C. M. Dall, Blaiotta, E., Omena, J. J., MacDonald, M., Bharati, S., & Smale, S. (2018). Objectionable queries. Searching for porn in app stores. In A. Helmond et al. (Eds.) Stores and their bias: repurposing ‘App Relatedness’?. DMI Summer School 2018. Retrieved from https://wiki.digitalmethods.net/Dmi/SummerSchool2018AppStoresBias

Jacomy, M., Venturini, T., Heymann, S., & Bastian, M. (2014). ForceAtlas2, a Continuous Graph Layout Algorithm for Handy Network Visualization Designed for the Gephi Software. PloS One (9)6, e98679.doi:https://doi.org/10.1371/journal.pone.0098679 

Omena, J.J. (2019). Reading Digital Networks: Climate Emergency, Bolsonaro & Bot Image Circulation by Vision API. The social platforms. Retrieved from https://thesocialplatforms.wordpress.com/2019/12/07/reading-digital-networks/

Omena, J.J., & Amaral, I. (2019). Sistema de leitura de redes digitais multiplataforma. In J. J. Omena (Ed.), Métodos Digitais: Teoria-Prática-Crítica. Lisboa: ICNOVA.

Venturini, T., Jacomy, M., & Jensen, P. (2019). What do we see when we look at networks an introduction to visual network analysis and force-directed layouts. arXiv:1905.02202 [physics.soc-ph] Retrieved from https://arxiv.org/abs/1905.02202

Venturini, T., Jacomy, M., & Pereira, D. (2015). Visual Network Analysis. SciencesPo Media Lab working paper. Retrieved from https://www.researchgate.net/publication/278030230_Visual_Network_Analysis

Acknowledgement
We thank Inês Amaral for initially contributing to this project pitch proposal, particularly with the subject of Gender Studies and the idea of studying the representations of Masculinity across platforms.