How did I not read about story generation systems before?! My dad normally points out linguistic articles in New Scientist for me, but he missed this one.
From how I’ve seen programs, such as the computer program in the Turing test, handle semantics before, I’m quite impressed by the extracts in the New Scientist article. The article is about various computer programs attempts at writing ‘stories’. Often it seems programs strongly rely on collocations ( “a sequence of words or terms that co-occur more often than would be expected by chance”), but just because ‘merry’ is often found next to ‘Christmas’, it doesn’t always, and if a program is working on a ‘most of the time’ basis, there are going to be times when the program gets it wrong. The stories here seem, for the most part, to avoid this. I’d like to see the ‘stories’ that didn’t get selected for publication and find out a bit more about how the programs work. While halting, I was quite impressed by the seeming creativity.