opennlp.grok.preprocess.cattag
Class CatterME

java.lang.Object
  |
  +--opennlp.grok.preprocess.cattag.CatterME
All Implemented Interfaces:
opennlp.maxent.Evalable, opennlp.common.preprocess.Pipelink, opennlp.common.preprocess.POSTagger
Direct Known Subclasses:
EnglishCatterME

public class CatterME
extends java.lang.Object
implements opennlp.maxent.Evalable, opennlp.common.preprocess.POSTagger

Tags words with a category (from categorial grammar) based on their surrounding contexts and a maximum entropy model. This is very similar to "supertagging" from work in the Tree Adjoining Grammar tradition (e.g. Bangalore's U. of Pennsylvania Ph.D.).

Version:
$Revision: 1.7 $, $Date: 2002/02/08 12:13:36 $
Author:
Gann Bierner

Field Summary
protected  opennlp.maxent.ContextGenerator cg
          The feature context generator.
protected  opennlp.maxent.MaxentModel model
          The maximum entropy model to use to evaluate contexts.
 
Constructor Summary
protected CatterME()
           
  CatterME(opennlp.maxent.MaxentModel mod)
           
 
Method Summary
static java.util.List bestSequence(java.util.List words, opennlp.maxent.MaxentModel model, opennlp.maxent.ContextGenerator cg)
           
 opennlp.maxent.EventCollector getEventCollector(java.io.Reader r)
           
 java.lang.String getNegativeOutcome()
           
 void localEval(opennlp.maxent.MaxentModel model, java.io.Reader r, opennlp.maxent.Evalable e, boolean verbose)
           
static void main(java.lang.String[] args)
           
 void process(opennlp.common.xml.NLPDocument doc)
          Label the words in a document with categories.
 java.util.Set requires()
           
 java.util.List tag(java.util.List sentence)
           
 java.lang.String tag(java.lang.String sentence)
           
 java.lang.String[] tag(java.lang.String[] sentence)
           
 
Methods inherited from class java.lang.Object
clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait
 

Field Detail

model

protected opennlp.maxent.MaxentModel model
The maximum entropy model to use to evaluate contexts.


cg

protected final opennlp.maxent.ContextGenerator cg
The feature context generator.

Constructor Detail

CatterME

protected CatterME()

CatterME

public CatterME(opennlp.maxent.MaxentModel mod)
Method Detail

getNegativeOutcome

public java.lang.String getNegativeOutcome()
Specified by:
getNegativeOutcome in interface opennlp.maxent.Evalable

getEventCollector

public opennlp.maxent.EventCollector getEventCollector(java.io.Reader r)
Specified by:
getEventCollector in interface opennlp.maxent.Evalable

process

public void process(opennlp.common.xml.NLPDocument doc)
Label the words in a document with categories.

Specified by:
process in interface opennlp.common.preprocess.Pipelink

tag

public java.util.List tag(java.util.List sentence)
Specified by:
tag in interface opennlp.common.preprocess.POSTagger

tag

public java.lang.String[] tag(java.lang.String[] sentence)
Specified by:
tag in interface opennlp.common.preprocess.POSTagger

tag

public java.lang.String tag(java.lang.String sentence)
Specified by:
tag in interface opennlp.common.preprocess.POSTagger

localEval

public void localEval(opennlp.maxent.MaxentModel model,
                      java.io.Reader r,
                      opennlp.maxent.Evalable e,
                      boolean verbose)
Specified by:
localEval in interface opennlp.maxent.Evalable

bestSequence

public static java.util.List bestSequence(java.util.List words,
                                          opennlp.maxent.MaxentModel model,
                                          opennlp.maxent.ContextGenerator cg)

requires

public java.util.Set requires()
Specified by:
requires in interface opennlp.common.preprocess.Pipelink

main

public static void main(java.lang.String[] args)
                 throws java.io.IOException
java.io.IOException


Copyright © 2003 Jason Baldridge and Gann Bierner. All Rights Reserved.