The study also found that formally worded messages reduced these posts more intensely.  (Photo: Unsplash)

Personalized warnings that alert users to the consequences of posting hateful messages on Twitter reduce the number of such posts, according to a recent report published by the University of Cambridge (UK).

YOU CAN SEE: Twitter | Half-read tweets no longer disappear with the change introduced in the web version of the social network

Directed by Mustafa Mikdat Yildirim, a PhD candidate at the Center for Social Media and Politics at New York University and author of ‘Short of Suspension: How Suspension Warnings Can Reduce Hate Speech on Twitter’, the study concludes that personalized warnings could reduce hate speech on this social network.

As collected Engadget, researchers have identified the accounts that were at risk of being suspended for violating the anti-hate speech social network rules. To do this, they looked for people who had used at least one word that fit this type of crime and who also followed at least one account recently suspended for the use of that language.

YOU CAN SEE: Elon Musk: the singular question he asked on Twitter that many do not stop commenting on

The researchers created different test accounts to send notices to these individuals regarding the content they posted on their profiles. These messages mentioned that hate speech put these users at risk of being suspended and that this had already happened to someone who was among their followers.

“The user ‘@cuenta’ that you follow has been suspended and it is possible that this is due to the hate language that he has used”says one of the writings that appears in the study. “If you keep using hate speech, you could be temporarily suspended.”, write another message.

YOU CAN SEE: Improved iPhone privacy causes billions of dollars in losses to Facebook, Twitter and Google

According to the aforementioned medium, Yildirim’s team created accounts that identified themselves as professional researchers to send these notifications in order to be “as credible and convincing as possible”said the doctor.

Among some of the results of this research, it is pointed out that the warnings were effective, at least in the short term. “Our results show that a single warning tweet sent by an account with no more than 100 followers can decrease the proportion of hate language tweets by up to 10%, Engadget quotes, according to the official document.

In addition, this team of researchers has verified how the messages that were written in a more formal language reduced publications of these characteristics in a greater proportion. For Yildirim and the co-authors of the report, if these warnings came from the social network itself or from official organizations, negative publications could be avoided to a greater extent.

“What we have seen from this experiment is that the important thing is how we let these users know that an account or entity is monitoring and monitoring their behavior, is mentioned in the conclusions section.

This also indicates that “The fact that the use of these hate speech is seen by other people could be the most important factor why some users stopped posting messages of this type”.

The study also found that formally worded messages reduced these posts more intensely. (Photo: Unsplash)


How much has social media consumption grown during the pandemic?  - LPD
Given the global fall of Facebook, Instagram and WhatsApp two weeks ago, it is worth asking ourselves what happened to the consumption of social networks during the pandemic in Peru and why this fall affected us so much. To begin with, according to Comscore, 72% of Peruvians’ interactions on social networks occur on Facebook and 22% on Instagram. In addition, the consumption of social networks had a significant growth during the health crisis, according to a report by Kantar Insights.


Follow us on twitter:

Leave a Reply