The Neiman Journalism Lab ran a piece yesterday by Sam Petulla on the increasing use of "sentiment analysis" in journalism. If the term is unfamiliar, the practice probably isn't. It's the use of such things a Facebook comments and analytical software to separate positive comments from negative ones, and to see how they correlate with, say, the results of an election.
Petulla doesn't do much in the way of explaining how such analysis is conducted, but he does show that its use is increasing, and that it offers both promise and peril. "What’s interesting about the use of sentiment analysis by journalists, though, is that so many of the industry’s ongoing concerns seem to crystallize in its promise: how to deal with social media platforms where the power to publish now belongs to millions; how to find a way to speak more authoritatively about the world it reports on; and how to take complex questions and display them simply and visually," Petulla writes.
On the other side, he quotes Micah Sifry, editorial director of Personal Democracy Media, who tells him, "Whenever you hear the words 'sentiment analysis,' your BS detector should go up."
Petulla recounts Politico's analysis of Facebook comments on Republican candidates during presidential primary season to see whether the negativity in the comments could predict the outcome of primaries. It's intriguing, but Petulla links to a post by Sifry that says such analysis will "generate 'bogus' results."
According to Christopher Potts, a Stanford linguist, this kind of analysis depends upon machines that can identify the sentiments in phrases such as, "It is a great movie if you have the taste and sensibilities of a five-year-old boy." Is that a positive comment--a recommendation from one parent to another? Or a negative one, meaning "don't waste your time?"
We will undoubtedly be seeing more of this, and it might be used to gauge sentiment on, say, climate change, or electric vehicles, or gun control.
Be curious, but be wary.