that paper's new to me. Our work combines Miller&Guinness on perceptrons and entities with Cramer on passive agressive, plus some secret sauce. I'm not in a position to open source it just now, but that may change.
On Thu, Jan 14, 2010 at 2:12 PM, Olivier Grisel <[email protected]> wrote: > 2010/1/14 Benson Margulies <[email protected]>: >> >> If there's one NLP thing I know something about, now, it is named >> entity extraction with averaged perceptrons and passive-aggressive >> training. This has the advantage of being mathematically trivial >> unless you want to prove that it works, which is as about as useful as >> proving that bumblebees can (or can't) fly. > > This soounds very interesting. > > A quick googling gave me the following paper: > > A New Perceptron Algorithm for Sequence Labeling with Non-local Features > by Jun’ichi Kazama and Kentaro Torisawa > > http://www.aclweb.org/anthology/D/D07/D07-1033.pdf > > Any other pointer to a paper or sample open source code? > > -- > Olivier > http://twitter.com/ogrisel - http://code.oliviergrisel.name >
