edu.stanford.nlp.parser.lexparser

Class TreeCollinizer

java.lang.Object

edu.stanford.nlp.parser.lexparser.TreeCollinizer

All Implemented Interfaces:

AbstractCollinizer

public class TreeCollinizer
extends java.lang.Object
implements AbstractCollinizer

Does detransformations to a parsed sentence to map it back to the standard treebank form for output or evaluation. This version has Penn-Treebank-English-specific details, but can probably be used without harm on other treebanks. Returns labels to their basic category, removes punctuation (should be with respect to a gold tree, but currently isn't), deletes the boundary symbol, changes PRT labels to ADVP.

Author:

Dan Klein, Christopher Manning

Constructor Summary

Constructors

Constructor and Description
TreeCollinizer(TreebankLanguagePack tlp)
TreeCollinizer(TreebankLanguagePack tlp, boolean deletePunct, boolean fixCollinsBaseNP)
TreeCollinizer(TreebankLanguagePack tlp, boolean deletePunct, boolean fixCollinsBaseNP, int whOption)

Method Summary

Modifier and Type	Method and Description
java.lang.String	toString()
Tree	transformTree(Tree guess, Tree gold)

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, wait, wait, wait

Constructor Detail

TreeCollinizer

public TreeCollinizer(TreebankLanguagePack tlp)

TreeCollinizer

public TreeCollinizer(TreebankLanguagePack tlp,
                      boolean deletePunct,
                      boolean fixCollinsBaseNP)

TreeCollinizer

public TreeCollinizer(TreebankLanguagePack tlp,
                      boolean deletePunct,
                      boolean fixCollinsBaseNP,
                      int whOption)

Method Detail

toString

public java.lang.String toString()

Overrides:

toString in class java.lang.Object

transformTree

public Tree transformTree(Tree guess,
                          Tree gold)

Specified by:

transformTree in interface AbstractCollinizer