Can algorithms really tackle the fake news fiasco?
According to a study by the University of Oxford, the 2016 US presidential elections were a “watershed moment” with rampant fake news all over the social media. Both Trump and Clinton camps exploited big data to manipulate the public opinion. As can be easily imagined, this phenomena is not limited to just the election; this is happening right now as well. In order to keep the fake news from reaching the top of the web search, companies are developing and updating their AI-based algorithms.
Before going into the individual companies’ algorithm regarding the fake news, it is important to understand the concept of PageRank. PageRank is the dominant mode of endorsement; for instance, a page is ranked high if it is cited by other important pages. Thus, this mode of endorsement forms the basis for the PageRank measure of importance.
Google is one of the leading companies that try to combat the fake news. Its PageRank algorithm ranks the search results based on the pages that are linked to the publication. Because PageRank applies across multiple platforms, we can identify fake news from real.
Unlike most cases where AI is more dependable, in the case of combating fake news, human intervention is reliable because algorithms can be easily manipulated and cannot make complex editorial decisions. Thus, companies that deal with big data such as Google, Facebook and Yahoo! should develop a better algorithm for their PageRank system.
http://www.computerworld.in/feature/can-algorithms-really-tackle-fake-news-fiasco