5 ms·

There were about 400,000 bots posting political messages during the 2016 U.S. presidential election on Twitter, according to a research paper by Emilio

There were about 400,000 bots posting political messages during the 2016 U.S. presidential election on Twitter, according to a research paper by Emilio Ferrara, an assistant professor at the University of Southern California.

dmix 9y ago
The next line which you left out: > He told Bloomberg that he has discovered that the same group of 1,600 bots tweeting extremist right-wing posts in the U.S. elections also posted anti-Macron sentiment during the French elections and extremist right-wing content during the German elections this year. So of those 400k political bots, 1600 were identified as talking about fringe topics. Not to mention it's not clear how this fits into the bigger picture? What percentage of ALL of Twitter accounts are bots? How many of those bots are active? How many actually have an audience of real people, not just other bots? For ex: 400k fake accounts, with few real followers, out of hundreds of millions of real users talking politics may not seem as significant... Not saying it's not a 'problem', but it needs to be understood in context, especially considering the constant news coverage and outrage.
- boomboomsubban 9y ago
  >So of those 400k political bots, 1600 were identified as talking about fringe topics. Read it again. Of those 400k bots, 1600 were identified talking about the US elections, anti-Macron sentiment, and extremist far-right topics. All three topics. That could be one person/group reusing bots. Additionally, hundreds of millions of real users talking politics is unlikely. Twitter has 300 milliom active users per month, 9-15% of them bots, the US election had 120 million voters. Hundreds of millions would be nearly everyone.
  - dmix 9y ago
    > Twitter has 300 milliom active users per month, 9-15% of them bots, the US election had 120 million voters Fair enough. So if 10% of voters used Twitter, that means 3% (400k) were bots, and of those a tiny percent were fringe/far-right.
    - boomboomsubban 9y ago
      >tiny percent were fringe/far-right. Nothing posted is proof of that. The 1600 number is interesting as it means one entity was trying to influence elections in three countries. It is not the number that were far-right, that number was not in this article. As context is supposed to be a needed thing, you could have just found the research paper rather than perform guestimates math. Fifteen percent of users tweeting about the election were bots, twenty percent of tweets, and about three Trump bots for every Hilary bot. http://firstmonday.org/ojs/index.php/fm/article/view/7090/5653 http://firstmonday.org/ojs/index.php/fm/article/view/7090/56...

dmix 9y ago

The next line which you left out: > He told Bloomberg that he has discovered that the same group of 1,600 bots tweeting extremist right-wing posts in the U.S. elections also posted anti-Macron sentiment during the French elections and extremist right-wing content during the German elections this year. So of those 400k political bots, 1600 were identified as talking about fringe topics. Not to mention it's not clear how this fits into the bigger picture? What percentage of ALL of Twitter accounts are bots? How many of those bots are active? How many actually have an audience of real people, not just other bots? For ex: 400k fake accounts, with few real followers, out of hundreds of millions of real users talking politics may not seem as significant... Not saying it's not a 'problem', but it needs to be understood in context, especially considering the constant news coverage and outrage.

boomboomsubban 9y ago
>So of those 400k political bots, 1600 were identified as talking about fringe topics. Read it again. Of those 400k bots, 1600 were identified talking about the US elections, anti-Macron sentiment, and extremist far-right topics. All three topics. That could be one person/group reusing bots. Additionally, hundreds of millions of real users talking politics is unlikely. Twitter has 300 milliom active users per month, 9-15% of them bots, the US election had 120 million voters. Hundreds of millions would be nearly everyone.
- dmix 9y ago
  > Twitter has 300 milliom active users per month, 9-15% of them bots, the US election had 120 million voters Fair enough. So if 10% of voters used Twitter, that means 3% (400k) were bots, and of those a tiny percent were fringe/far-right.
  - boomboomsubban 9y ago
    >tiny percent were fringe/far-right. Nothing posted is proof of that. The 1600 number is interesting as it means one entity was trying to influence elections in three countries. It is not the number that were far-right, that number was not in this article. As context is supposed to be a needed thing, you could have just found the research paper rather than perform guestimates math. Fifteen percent of users tweeting about the election were bots, twenty percent of tweets, and about three Trump bots for every Hilary bot. http://firstmonday.org/ojs/index.php/fm/article/view/7090/5653 http://firstmonday.org/ojs/index.php/fm/article/view/7090/56...