It struck me today as I wrote this tweet:
..that I was expressing a negative sentiment towards a product that would not only be missed by an automated sentiment analysis program, but also by a social media monitoring platform, period. Since I didn't name the product in any of my previous or subsequent tweets (and won't here, either--this isn't about them), any kind of automated monitoring would have to go through Amber Naslund's tweet stream and essentially intuit which of her tweets I was responding to.
Obviously, I'm not the first person to notice this. But I am curious: setting aside the sentiment analysis problem, is anyone working strictly on the content monitoring problem? A human would look at the time of my tweet, start there with Amber's tweets, and work backward until an assumption made sense. Is anyone in the social media monitoring space thinking along these lines?