Figure 1
Non-directed network (left) and directed network (right).
Figure 2
Ideal example of three different group formations performing a bipartite network analysis: a) the orange group exists because its members (i.e., authors 3, 5, and 9) share the same representational contents (i.e., Topic 1); b) the green group exists because its members (Authors 7, 8, and 10) interact; c) the purple group exists because its members (Authors 1, 2, 4, and 6) both share the same representational contents (Topic 2 and 3) and interact with each other.
Note: Each colour represents a group based on modularity class. The dimension of the nodes’ labels is proportional to the in-degree.
Figure 3
Dendrogram of the topics resulting from the Reinert method. English translation by the authors.
Figure 4
Bipartite network analysis output of the example of application (cf. graph file in the supplementary material). The thickness of the arc is determined by their weight. The size of the node labels is proportional to the in-degree. Each community is represented by a different colour and is defined by the modularity class.
Table 1
List of profiles with the highest indegree in the network.
| RANKING | ID | PROFILE | ACTIVITY | INDEGREE |
|---|---|---|---|---|
| 1 | 308 | @jairbolsonaro | President of Brazil | 8.0 |
| 2 | 314 | @jdoriajr | Governor of Sao Paulo | 7.0 |
| 3 | 247 | @folha | Newspaper | 4.0 |
| 4 | 311 | @JanainaDoBrasil | São Paulo State Representative | 4.0 |
| 5 | 90 | @BlogdoNoblat | Journalist | 3.0 |
| 6 | 257 | @GabrielaPrioli | Lawyer and Political Commentator | 3.0 |
| 7 | 615 | @XicoGraziano | Politician | 3.0 |
| 8 | 260 | @Gavin07290309 | --- | 3.0 |
| 9 | 40 | @AnaPaulaVolei | Political Commentator | 2.0 |
| 10 | 61 | @ArthurWeint | Minister of Education (Bolsonaro Government) | 2.0 |
Figure 5
The figure shows the distribution of community sizes by modularity class, highlighting classes 0, 8, 10, 12, 15, and 22. These classes represent communities formed by nodes connected to topics 1 through 6, identified using the Reinert Method. Each of these nodes is linked to at least one of the six topics. Other modular groups, which are not related to these topics, only represent ‘user-user’ relationships.
Figure 6
Multiple User-Topic and User-User connections.
Note: Zooms of Figure 4; labels A, B, C, D, and E – representing users – have been manually added for clarity of exposition.