Regional patterns and networked behavior: interdisciplinary methods for identifying bots in the 2020 elections in Brazil

Authors

  • João Guilherme Bastos dos Santos Instituto Nacional de Ciência e Tecnologia em Democracia Digital (INCT.DD) / Rooted in Trust (RiT)/Internews
  • Arthur Ituassu Pontifícia Universidade Católica do Rio de Janeiro (PUC-RJ)
  • Sérgio Lifschitz Pontifícia Universidade Católica do Rio de Janeiro (PUC-RJ)
  • Yago Cury Pontifícia Universidade Católica do Rio de Janeiro (PUC-RJ)
  • Thayane Guimarães Universidade Federal Fluminense (UFF)
  • Diego Cerqueira Universidade Federal do Rio de Janeiro (PESC/COPPE - UFRJ)
  • Debora Albu London School of Economics
  • Redson Fernando Instituto de Tecnologia e Sociedade (ITS Rio)
  • Julia Hellen Ferreira Universidade Federal Fluminense (UFF)
  • Maria Luiza Mondelli Laboratório Nacional de Computação Científica (LNCC)

DOI:

https://doi.org/10.5753/isys.2022.2415

Keywords:

Complex networks, Network science, Elections, Social bots

Abstract

To improve the detection of automated behavior on social networks in the Brazilian scenario, this work applies five methods of bot identification - analysis of (i) networks, (ii) sentiment, (iii) profile, (iv) temporal, and (v) crossing of these analyses - to find possible bots posting on the comments of tweets from candidates for the mayors' election in the capitals Rio de Janeiro, São Paulo, Recife, Porto Alegre and Fortaleza in 2020. After identification, we followed two strands of analysis (namely, network and lexical). We seek to better understand if there are different types of bots in each region, different network performances and vocabularies. Among the preliminary results obtained, we highlight the multiplicity of types of automation, understanding of regional variations and the role of bots in the Brazilian electoral scenario.

Downloads

Download data is not yet available.

References

Almeida, Y. L., Rubin, F. S., de Faria Alvim, A. C., Dias, V. M. F., and dos Santos, R. P. (2020). O uso das redes sociais para interferir nas democracias: Um mapeamento sistemático da literatura. In Anais do IX Brazilian Workshop on Social Network Analysis and Mining, pages 178–183. SBC.

Bouzy, C. (2018). Botsentinel: link.

Leite, M., Guelpeli, M., and Santos, C. (2020). Um modelo baseado em regras para a detecção de bots no twitter. In Anais do IX Brazilian Workshop on Social Network Analysis and Mining, pages 37–48, Porto Alegre, RS, Brasil. SBC.

Rodriguez, A., Sava, P. S., Ituassu, A., and Lifschitz, S. (2019). Sistema web crawler para coleta automática de tweets, persistência em bancos de dados e análises estatı́sticas. In Proceedings Companion 34o Simpósio Brasileiro de Bancos de Dados (SBBD), 1o Workshop de Trabalhos de Alunos de Graduação (WTAG), pages 325–332, Fortaleza, CE, Brasil.

Rofrı́o, D., Ruiz, A., Sosebee, E., Raza, Q., Bashir, A., Crandall, J., and Sandoval, R. (2019). Presidential elections in ecuador: Bot presence in twitter. In 2019 Sixth International Conference on eDemocracy eGovernment (ICEDEG), pages 218–223.

Ruediger, M. A., Grassi, A., Freitas, A., Contarato, A. d. S., Taboada, C., Carvalho, D., Ferreira, H., Silva, L. R. d., Lenhard, P., Bastos, R., et al. (2017). Robôs, redes sociais e polı́tica no brasil: estudo sobre interferências ilegı́timas no debate público na web, riscos à democracia e processo eleitoral de 2018.

Santos, J. G., Ituassu, A., Lifschitz, S., Guimarães, T., Cerqueira, D., Albu, D., Fernando, R., Ferreira, J., and Mondelli, M. L. (2021). Das milı́cias digitais ao comportamento coordenado: métodos interdisciplinares de análise e identificação de bots nas eleições brasileiras. In Anais do X Brazilian Workshop on Social Network Analysis and Mining, pages 187–192, Porto Alegre, RS, Brasil. SBC.

Varol, O., Ferrara, E., Davis, C., Menczer, F., and Flammini, A. (2017). Online human-bot interactions: Detection, estimation, and characterization. In Proceedings of the international AAAI conference on web and social media, volume 11.

Woolley, S. C. and Howard, P. N. (2018). Computational propaganda: Political parties, politicians, and political manipulation on social media. Oxford University Press.

Published

2022-12-30

How to Cite

Bastos dos Santos, J. G., Ituassu, A., Lifschitz, S., Cury, Y., Guimarães, T., Cerqueira, D., Albu, D., Fernando, R., Ferreira, J. H., & Mondelli, M. L. (2022). Regional patterns and networked behavior: interdisciplinary methods for identifying bots in the 2020 elections in Brazil. ISys - Brazilian Journal of Information Systems, 15(1), 26:1–26:12. https://doi.org/10.5753/isys.2022.2415

Issue

Section

Extended versions of selected articles