After some delay, I can announce that the LibraryThing/uClassify contest has been won by Kelly Vista—the only entrant, but a worthy one. (Kelly gets a copy of Programming Collective Intelligence and $100 from Amazon or IndieBound.) She described her “LibraryThing classifier” as follows:
“My goal was to create a classifier that would automatically “tag” any book description based on actual LibraryThing tags. For example, if you paste the book description for “Truman” into UClassify, it should return to you LibraryThing tags that suit the book. This is one step more general than one of [Tim’s] ideas (fiction vs. non-fiction).”
In my testing, it does a pretty good job of hitting the top tags. Pasted descrptions of Harry Potter give “young adult” and “children’s.” John Adams gives “american history” and “biography.” It’s not perfect—Adams is also labelled “young adult”—but the initial results are good and the whole point of uClassify is to enable accelerating accuracy.
uClassify seems to be growing apace. They recently opened up public classifications for external access, so I’ll be looking into automatic text-language classification of LibraryThing reviews.