This article (13 page PDF) introduces "a novel approach to probe media diet models – language models adapted to online news, TV broadcast, or radio show content – that can emulate the opinions of subpopulations that have consumed a set of media." According to the authors, it can predict how a bdy of media content will change public opinion. If these findings can be validated, they tell us two things: first, we can predict ahead of time what the public will feel about something, and second, that media (including cable news, advertising, social media, and the rest) does have an impact on public opinion (even if it's not a nice neat cause-and-effect relation). What would we do with this information: use it to manipulate people, or exercise limits on media? I'm not sure there's a third option or middle ground here.
Today: 6 Total: 108 [Direct link] [Share]
