Pagerank and the spread of Fake News
Ever since the 2016 elections, there has been a large public debate about “Fake News” and the role of companies like Facebook and Google in its spread on the web. The story typically follows the same pattern: a fake news article is shared immensely on social media, often initially by bot accounts, reaching a “trending” status, tricking people into thinking it is legitimate. We have had lethal examples like “Pizzagate” as well as false information spread about the identity of the Las Vegas shooter that started arising right after the event.
The spread of such articles can be studies by analyzing the network, as its impact is proportional with the number of people it reaches. Combined with this is the fact that the Pagerank algorithm that Google makes that the most popular articles show up first, creating a snowball effect on the internet. It is for this reason that Google (and Facebook) are increasingly criticized for helping spread fake news. Although Google claims it is using a more complex version of Pagerank to filter out the false information, this opened up a debate on the “accomplice” role they play in this story, as well as a debate on censorship on the internet. The article mostly focuses on their newer “Top Stories” feature, which showed a fake 4chan post after the Las Vegas incident. As a side note, it is far from the first time 4chan has been involved in controversial internet events like these.
It will be interesting to see how this develops towards the future. A few questions are waiting to be answered. Will AI become better at spotting fake news? Will the internet become more censored? Will websites like Google, Facebook, Twitter and 4chan be help accountable for spreading fake news?