by Alessandra Maia Terra de Faria, Carlos Trucíos, and Marcelos Cantañeda de Araújo

Reviewed by Matheus Lucas Hebling

Find out what and how presidential candidates tweet and are tweeted.



This research aimed to follow the tweets related to the main presidential candidates according to the opinion polls available for the 2022 elections in Brazil.



Daily tweets spanning from October 3rd to October 29th of 2022 (the day after the second turn elections in Brazil) were collected for each one of the candidates running in the Brazilian second turn presidential elections. Tweets were collected from both candidates’ timelines and Twitter users mentioning the candidates, totaling more than 69 million tweets, the largest volume of monthly data obtained since the beginning of the survey. Data were extracted through a Twitter API used exclusively for academic purposes and analyzed using R software.

The authors thank Twitter for the academic accounts granted to them.

Herein is the updated data (October versus September, data as of November 29th, 2022) of Twitter followers for each of the candidates.

  • Bolsonaro – from 10.5 up to 11.2 million (about 6.7% of the increase in followers in comparison to the previous month)

  • Lula – from 6.3 up to 7.5 million (about 19% of the increase in followers in comparison to the previous month)

Candidates’ tweets

In Figure 1, we report the number of tweets on the candidates’ timelines, among the last two running for the second turn that was part of our survey: Lula and Bolsonaro, according to the frequency with which the candidates tweeted in October.

Figure 1: Timelines

For the first time, we observed a considerable distance between the percentages of the manner in that candidates positioned themselves on the social network. Lula da Silva increased his activity substantially and Jair Bolsonaro decreased his interactions.

Figures 2 and 3 present the most frequent words in the candidates’ timeline tweets and the most frequent words in the candidates’ timeline tweets weighted by the inverse document frequency (TF-IDF), respectively.

Figure 2: Most frequently used words in the candidates’ timeline.

The analysis of the most frequent words in candidates’ timeline tweets (Figure 2) allows us to present a dominant scenery of subjects they deal with. In common, both profiles present the term “people” [“povo”] which was a novelty in September and was kept in the political public scenario of Twitter through October, in addition to “Brasil”, “Lula”, “Bolsonaro” and “country” [“país”]. If we observe Bolsonaro’s profile itself, the terms “bigger” [“maior”], “path” [“caminho”], “million” [“milhões”], “2022” and “against” [“contra”] stood out. Comprehensively, in Lula’s profile, the emphasis remained on the verb “to do” [“fazer”], now accompanied by the terms “folks” [“gente”], “day” [“dia”], “government” [“governo”] and the explanatory conjunction “because” [“porque”].

Figure 3 TF-IDF by candidates’ timeline

In Figure 3, the TF-IDF (term frequency-inverse document frequency) reflects the frequency of words in candidate timeline tweets that are infrequent for the candidates. Thereby:

  • In Lula’s profile, the verbs “to vote” [“votar”] and “know” [“sabe”] stand out, as well as the terms “president” [“presidente”], “universities” [“universidades”] and “hunger” [fome].

  • It is remarkable that Bolsonaro’s profile presents no verbs. The terms that appear are “path” [“caminho”], “2022”, “reduction” [“redução”], “corruption” [“corrupção”], and “drugs” [“drogas”].

Tweets about the candidates

The total number of tweets mentioning each candidate is displayed in Figure 4 and the daily evolution in Figure 5.

Next, in Figure 4, we present, in descending order (from the most cited to the least cited), the total number of tweets that mentioned the name of each candidate surveyed in the month of October: Bolsonaro and Lula.

To collect the tweets mentioning the respective candidates, the words “Bolsonaro” and “Lula” were used as search criteria.

Figure 4: Number of tweets mentioning the candidates.

The volume of tweets about the candidates more than doubled. To give you an idea, in September the number of tweets mentioning Bolsonaro was 16,784,820 and Lula’s was 13,899,696.

The daily evolution of tweets mentioning each candidate is shown in Figure 5. We can observe that although Jair Bolsonaro remained in the lead for most of the second round, if we consider interactions on the social network, this fact did not make his advantage overwhelming, especially when we notice the number of followers that is largely favorable to him at that time and until the present day when compared to the opponent, then-candidate Lula da Silva, later elected president.

Figure 5: Daily evolution of tweets mentioning the candidates.

Word clouds 

Finally, we present below both word clouds with, excluding stop words, the top 100 words used in the interactions of Twitter users in October. For better visualization, each candidate’s name was taken from its cloud.

A word cloud is a graphical representation of the most frequent words within a text or set of texts.

When analyzing word clouds, each one corresponds to a candidate. It is important to point out that each candidate’s name was taken from its cloud, for better visualization of the associated words. It should also be noted that each cloud reflects the 100 most relevant words associated, excluding stop words, to each candidate in the interactions of Twitter users from the third day until the twenty-ninth of October.

In text analysis, stop words are quite common words such as “and”, “from”, “the”, etc. These words are not useful for analysis and are often removed before analysis.

Figure 6: Word cloud for Bolsonaro

Figure 7: Word cloud for Lula

When analyzing the clouds, we share the first impression of each one:

  • Bolsonaro: “Lula” appears in the foreground alone. In the background, we have the words “president” [“presidente”] and “Brasil”. Then “video”, “against” [“contra”], “in”, “government” [“governo”], and “now” [“agora”].

  • Lula: in the foreground, we see “president” [“presidente”], “Brasil” and the English preposition “in”. In the background stood out other English terms, that probably relate to the internalization of comments about Brazilian elections in 2022. We can observe the words “demand”, “withheld”, “Brazil” (with “z”), “PT”, “response”, “vote” [“voto”], “votes” [“votos”], “turn”, “has”, “account”, “learn”, “more”, “learn”, “legal” (The word legal in Portuguese has both possible meanings: as something “cool” and as something “juridical” as it means in English).

Sentiment analysis

The sentiment of each tweet was constructed by identifying the sentiments of the basic units (the words) using the Oplexicon v3.0 and Sentilex dictionaries, from the LexiconPT Package. Thus, each word found in the dictionaries receives 1, -1, or 0 scores, depending on whether the feeling is positive, negative, or neutral, respectively. Words not found in the dictionaries also receive a 0 score. The values assigned to each word within the tweet were added up, and depending on the result positive, negative, or zero, the sentiment of the tweet is classified. In Figure 8, feelings (Negative, Neutral, and Positive) are presented in percentages per candidate. It is possible to highlight a balance between the feelings expressed in the tweets of the three candidates. Such data will be monitored over time comparatively. This is a picture, a sentimental snapshot from the 03rd until the 29th of October on Twitter.

Candidate Lula obtained a higher percentage of positive and neutral feelings compared to candidate Jair Bolsonaro, who obtained the highest percentage of tweets classified as negative comparatively (31.49%).

Figure 8: Sentiments of tweets per candidate

Next, it will be possible to look at the word cloud of each candidate, separately, according to the feelings attributed to each tweet, in Figures 9 and 10. Words in pink appear in tweets rated as associated with positive feelings, words in blue appear in tweets rated as associated with negative feelings, and words in beige appear in tweets rated as neutral. The word clouds are considered the 200 most frequent words.

Figure 9: Word cloud Sentiments for Bolsonaro

Figure 10: Word cloud Sentiments for Lula

  • Bolsonaro: Tweets related to the candidate Bolsonaro that were classified as associated with positive sentiments are characterized by words such as “in”, “learn”, “Brazil”, “response”, “demand”, “account”, “more”, “to”, “withheld”, “has”, “been”. Tweets classified as associated with negative feelings are characterized by words such as “to vote” [“votar”], “Roberto Jefferson”[1], “bad guy” [“bandido”], “dude” [“cara”]. Finally, tweets considered neutral highlight “move” [“mexa”], “salary” [“salário”], “Janones[2]”, “Lula”, “22”, “censored” [“censurou”].

  • Lula: Tweets related to candidate Lula that were classified as associated with positive feelings are characterized by words such as “in”, “learn”, “been”, “legal”, “demand”, “account”, “response”, “more”, “has”, “to”, “withheld”, “Brazil”. The tweets classified as negative are characterized by “to vote” [“votar”], followed by “corruption” [“corrupção”], “dude” [“cara”], and “convict” [“presidiário”]. Finally, tweets with neutral sentiment are mainly characterized by the terms “Bolsonaro” and “president” [“president”], followed by “turno” [“turn”] and “13”.

Final comments

The presentation of this dataset aimed to contribute to interpretations about the movement on Twitter of presidential candidates in the 2022 elections in Brazil, as well as about what is said about them in the interactions of users of the platform, considering a long-term analysis, with data collected from April to October 2022. This Report considers data until the last day after elections in October, in comparison to what was found in September. The material obtained before can be found and compared with former Reports of August, July, June, May, and  April.

[1] Known as “the sniper of Bolsonaro” (, Roberto Jefferson’s account on Twitter ( @bobjeffcensored ) was suspended in 2021, after what he was indicted and arrested for the first time. ( In 2022, by the moment opinion polls suggested that candidate Lula would finally defeat candidate Bolsonaro, former congressman Mr. Jefferson violated conditions of his house arrest, posting a video on social media verbally attacking a justice of the Supreme Court. As the Federal Police arrived at his home in the countryside of Rio de Janeiro to escort him back to jail, Mr. Jefferson welcomed them with dozens of shots from his military-grade assault rifle, as well as two grenades. The ensuing standoff was an unmitigated disaster for the Bolsonaro campaign, which seemed confused about how to react. (Read more about the incident at: The international press noticed by them that there was a 40% increase in violence against candidates during that moment in Brazil, with 140 attacks between July and September of 2022. That was considered one of the most bitter campaigns since the country’s return to democracy in 1985, marked by a series of brutal killings that police believe were politically motivated. (Find more about that topic at:

[2]Representative André Janones (@AndreJanonesAdv) is the son of a domestic worker and a wheelchair-bound father from a small city in the interior of Brazil. He paid his way through law school by working as a bus fare collector and only became a federal lawmaker in 2019. He joined forces with candidate Lula’s campaign in August, and he was the most prominent Lula ally to drop the gloves in a bruising run-off race that took even Bolsonaro’s campaign by surprise. (Read more about his influence at )

Alessandra Maia Terra de Faria, Social Sciences Department at PUC-RIO / PPGCS – UFRRJ. E-mail:

Carlos Trucíos, Department of Statistics, University of Campinas. E-mail:

Marcelo Castañeda de Araujo, Department of Business/UFRJ.

Alessandra Maia Terra de Faria, Carlos Trucíos and Marcelo Castañeda de Araujo (2023) "October 2022: Brazilian’s presidential candidates on Twitter". Brazilian Research and Studies Blog. ISSN 2701-4924. ISSN 2701-4924nameVol. 3 Num. 1. available at:, accessed on: February 27, 2024.