ah NLP tied into propositional logic. I remember that stuff in the 90s: I think MIT had a thing for it. It’s back when the letters AI were guaranteed death for funding. It was an AI winter, only because they rarely used the phrase… but they used other phrases instead, which were actually more accurate.
I’ve seen some good work done in sentiment analysis lately – lots of good code floating around for that. [I did some sentiment analysis in the late 90s/00s in Excel and VBA … didn’t know it was called that. I just got a CD full of people’s free-form opinions about the product (it was Claritin), and I had to somehow figure out how to derive their moods from it.
I did it: Actually wasn’t very hard at all, really. How good was it? Well, good enough for the bosses.