Is that really the standard term? You know, that the LW party line is that it’s a bad term like selling non-apples. Google suggests to me that it is not the most popular term. The link below replaces “non-traditional” with “modern,” which isn’t an improvement on this dimension.
Also, my first parsing was that “non-traditional” modified “authorship.” This is actually a reasonable use of the prefix “non,” since having a strong prior on the author makes a big difference (sociologically, if not technically). How bout that Marlowe?
You’re right, it’s a horrible term. For one thing, the methods involved are pretty well-established by now. I just use it by habit. As for that old Marlowe/Shakespeare hubbub, here’s a recent study which finds their style similar but definitely not identical.
I skimmed it and nothing seemed obviously wrong. If you’re interested, you could try for yourself. If you download Marlowe’s corpus, Shakespeare’s corpus and stylo you can get a feel for how this works in a couple of hours.
Is that really the standard term? You know, that the LW party line is that it’s a bad term like selling non-apples. Google suggests to me that it is not the most popular term. The link below replaces “non-traditional” with “modern,” which isn’t an improvement on this dimension.
Also, my first parsing was that “non-traditional” modified “authorship.” This is actually a reasonable use of the prefix “non,” since having a strong prior on the author makes a big difference (sociologically, if not technically). How bout that Marlowe?
You’re right, it’s a horrible term. For one thing, the methods involved are pretty well-established by now. I just use it by habit. As for that old Marlowe/Shakespeare hubbub, here’s a recent study which finds their style similar but definitely not identical.
Does anyone use a better term? “Statistical author attribution” seems like an obvious term, but google tells me that no one has ever used it.
Have you read the study you link? People who have read it tell me that the conclusions drawn do not match the body of the paper.
I skimmed it and nothing seemed obviously wrong. If you’re interested, you could try for yourself. If you download Marlowe’s corpus, Shakespeare’s corpus and stylo you can get a feel for how this works in a couple of hours.