Bow-tie structures of twitter discursive communities

Abstract

Bow-tie structures were introduced to describe the World Wide Web (WWW): in the direct network in which the nodes are the websites and the edges are the hyperlinks connecting them, the greatest number of nodes takes part to a bow-tie, i.e. a Weakly Connected Component (WCC) composed of 3 main sectors: IN, OUT and SCC. SCC is the main Strongly Connected Component of WCC, i.e. the greatest subgraph in which each node is reachable by any other one. The IN and OUT sectors are the set of nodes not included in SCC that, respectively, can access and are accessible to nodes in SCC. In the WWW, the greatest part of the websites can be found in the SCC, while the search engines belong to IN and the authorities, as Wikipedia, are in OUT. In the analysis of Twitter debate, the recent literature focused on discursive communities, i.e. clusters of accounts interacting among themselves via retweets. In the present work, we studied discursive communities in 8 different thematic Twitter datasets in various languages. Surprisingly, we observed that almost all discursive communities therein display a bow-tie structure during political or societal debates. Instead, they are absent when the argument of the discussion is different as sport events, as in the case of Euro2020 Turkish and Italian datasets. We furthermore analysed the quality of the content created in the various sectors of the different discursive communities, using the domain annotation from the fact-checking website Newsguard: we observe that, when the discursive community is affected by m/disinformation, the content with the lowest quality is the one produced and shared in SCC and, in particular, a strong incidence of low- or non-reputable messages is present in the flow of retweets between the SCC and the OUT sectors. In this sense, in discursive communities affected by m/disinformation, the greatest part of the accounts has access to a great variety of contents, but whose quality is, in general, quite low; such a situation perfectly describes the phenomenon of infodemic, i.e. the access to "an excessive amount of information about a problem, which makes it difficult to identify a solution", according to WHO.

Figure 1

The seven sectors of Yang’s bow-tie structure.

Figure 2

Percentages of nodes in each discursive community, Italian COVID-19 dataset. Due to the presence of politicians and political parties from a specific political area, the various discursive communities are called following their political alignment. “PD” stays for the Italian Democratic Party (Partito Democratico); Italia Viva (“IV”) is the political party of the former prime Minister and former PD secretary Matteo Renzi, while M5S is the “Movimento 5 Stelle”, a political movement born on the web and being the most represented party in the Italian parliament at the time of the data collection. “FI” stays for Forza Italia, the political party of the former Prime Minister Silvio Berlusconi, while the “DX” (Destra) community includes right wing parties as Lega and Fratelli d’Italia. The most crowded discursive community is the one of Media in which there are most of the online news outcasts and newspapers. The accounts for which it was not possible to assign a discursive community are in grey.

Figure 3

The bow-tie structure of the discursive communities in the Italian COVID-19 dataset. The dimension of the sectors is proportional to the number of nodes: DX and IV discursive communities have strong bow-ties (the OTHERS block is smaller than SCC), while the others are weak (the OTHERS block is greater than SCC, still being smaller then bow-tie WCC). The DX, IV, FI and MEDIA groups display a OUT-dominant bow-tie structure, with the most part of the nodes located in the OUT sector. M5S and PD communities have a INTEND-dominant bow-tie structure, the INTENDRILS sector being the dominant one. The colour of the blocks quantifies the distance between the observed dimensions and those predicted by the Direct Configuration Model (DCM). The observed dimension for the OTHERS sector is significantly less numerous (considering a significance level at $\alpha =0.01$) for all the communities, but PD. Remarkably, for INTEND-dominant bow-ties, also other sectors, as SCC and INTENDRILS, are usually bigger than what we expect from the model.

Figure 4

Distribution of the percentage of verified users in each sector of the discursive communities with, respectively, OUT-dominant, INTEND-dominant and not informative bow-ties. Each bar-chart displays the average percentage of verified users in a specific sector, calculated respectively for all the OUT-dominant, INTEND-dominant and not informative bow-ties. In the cases of OUT-dominant and INTEND-dominant bow-ties, the highest percentages of verified accounts can be found in the IN group. Moreover, in OUT-dominant bow-ties, we can found a relevant percentage of verified accounts also in the SCC. Naturally, for those communities with no bow-tie structure the verified accounts are mostly placed in the OTHERS sector and, to less extent, in the OUTTENDRILS one.

Figure 5

Percentage of verified accounts in the bow-tie sectors for each discursive community of the COVID-19 dataset. The bar-charts confirm that verified accounts are mainly located in the IN sector and, to a less extent, in the SCC one. Only for the PD group, which has a INTEND-dominant bow-tie structure, verified accounts are mostly placed in the INTENDRILS block.

Figure 6

Percentage of nodes and edges in SCC for the communities in the Italian COVID-19 dataset. In the Italian COVID-19 dataset, the conservative and right-oriented discursive community (DX) has more numerous and denser SCCs, as it is displayed in the highest two graphics. In the lowest graphic, it can be seen that, also considering the number of links per node in SCC, DX results again the first discursive community. These results hold for all the conservative groups in all the datasets under investigation.

Figure 7

Bow-tie structure of the DX group and percentages of retweets containing URLs of untrustworthy webpages. The DX community in the Italian COVID-19 dataset presents the highest number of retweets containing a link to untrustworthy webpages. Most of them origin from SCC: 8.4% of the retweets in SCC and 7.3% of the retweets between SCC and OUT contain not reliable URLs. In the diagram, we also insert the link between IN and OUT (the dashed line), which, considering the definition of each sector, is not forbidden a priori.

Figure 8

The bow-tie structure of the discursive communities for the Turkish EURO2020 dataset. The SPORTS group contains the official accounts of football players and clubs, and sports newspapers, while AK refers to the Justice and Development Party (Turkish: Adalet ve Kalkınma Partisi, AKP), which is a conservative political party in Turkey, including President Erdogan and his ministries. The SPORTS discursive community does not display an informative bow-tie structure, while the AK one has an extremely weak (INTEND-dominant) bow-tie. The dimension of the sectors is proportional to the number of nodes therein and the color quantifies the distance between the observed and the predicted dimension. Looking to the color of the vertices, it is possible to see that the observed dimensions are not statistically significant.

Figure 9

The bow-tie structure of the discursive communities for the Italian EURO2020 dataset. The dimension of the sectors is proportional to the number of nodes therein and the color quantifies the distance between the observed and the predicted dimension. The main discursive community is formed by football players, sports newspapers and journalists. Then, we identified a MEDIA community, containing accounts of Italian media, and three small political communities (DX, IV, M5S). MEDIA, DX and IV do not display an informative bow-tie structure (respectively 74%, 81.2% and 63.6% of the nodes in OTHERS), while FOOTBALLERS and M5S show a weak bow-tie (respectively 81.1% and 75.7% of nodes in INTENDRILS).

Figure 10

Percentages of social-bots in each sector of the bow-tie structure for OUT-dominant, INTEND-dominant and not informative bow-ties. This figure collects the percentage of bots in every bow-tie’s sector for discursive communities with OUT-dominant, INTEND-dominant and not informative bow-tie structure. It is easy to note how the highest percentages can be found in the OUT sector, in the INTENDRILS and in the OTHERS one, respectively for the case of OUT-dominant, INTEND-dominant and not informative bow-tie.