There is a coref component<http://incubator.apache.org/opennlp/documentation/1.5.2-incubating/manual/opennlp.html#tools.coref>, but it needs to be taken care of. See the discussion from November 2011<http://mail-archives.apache.org/mod_mbox/incubator-opennlp-dev/201111.mbox/browser>, named Coref problem (was: JWNL)<http://mail-archives.apache.org/mod_mbox/incubator-opennlp-dev/201111.mbox/%3C4EC4D73D.4000205%40gmail.com%3E> .
Aliaksandr P.S. By the way, a minor detail: in your example, there is no source to resolve *it* to, probably it was in a previous sentence. On Tue, Dec 27, 2011 at 5:28 AM, Khairul Fahmi <kahirulfa...@gmail.com>wrote: > I'm developing tool for text processing using OpenNLP. I want to know if > OpenNLP provide library to do 'anaphora resolution'. > I've read documentation provided by apache but none of them about anaphora > resolution. > What package should I use and what step I should do so that I can produce > output like the following sample? > > My sample sentence is "If *we* don't do *it* now, Australia is going to be > in deficit and debt into the next century." > I need to replace *we* and *it* in sentence to become "If *Australia* > don't do *spending cuts* now, Australia is going to be in deficit and debt > into the next century." > > > Thanks. > > Khairul Fahmi > >
