Package edu.stanford.nlp.trees.international.arabic

Class ArabicTreeNormalizer

java.lang.Object

edu.stanford.nlp.trees.TreeNormalizer

edu.stanford.nlp.trees.BobChrisTreeNormalizer

edu.stanford.nlp.trees.international.arabic.ArabicTreeNormalizer

All Implemented Interfaces:

TreeTransformer, Serializable, Function<Tree,Tree>

public class ArabicTreeNormalizer
extends BobChrisTreeNormalizer

Normalizes both terminals and non-terminals in Penn Arabic Treebank (ATB) trees. Among the normalizations that can be performed:

• Adds a ROOT node to the top of every tree
• Strips all the interesting stuff off of the POS tags.
• Can keep NP-TMP annotations (retainNPTmp parameter)
• Can keep whatever annotations there are on verbs that are sisters to predicatively marked (-PRD) elements (markPRDverb parameter) [Chris Nov 2006: I'm a bit unsure on that one!]
• Can keep categories unchanged, i.e., not mapped to basic categories (changeNoLabels parameter)
• Counts pronoun deletions ("nullp" and "_") as empty; filters

Author:

Roger Levy, Anna Rafferty, Spence Green

See Also:

Serialized Form

Nested Class Summary

Nested Classes

Modifier and Type	Class	Description
static class	ArabicTreeNormalizer.ArabicEmptyFilter	Remove traces and pronoun deletion markers.

Nested classes/interfaces inherited from class edu.stanford.nlp.trees.BobChrisTreeNormalizer

BobChrisTreeNormalizer.AOverAFilter, BobChrisTreeNormalizer.EmptyFilter

Field Summary

Fields inherited from class edu.stanford.nlp.trees.BobChrisTreeNormalizer

aOverAFilter, emptyFilter, tlp

Constructor Summary

Constructors

Constructor	Description
ArabicTreeNormalizer()
ArabicTreeNormalizer(boolean retainNPTmp)
ArabicTreeNormalizer(boolean retainNPTmp, boolean markPRDverb)
ArabicTreeNormalizer(boolean retainNPTmp, boolean markPRDverb, boolean changeNoLabels)
ArabicTreeNormalizer(boolean retainNPTmp, boolean markPRDverb, boolean changeNoLabels, boolean retainNPSbj, boolean retainPPClr)

Method Summary

Modifier and Type	Method	Description
String	normalizeNonterminal(String category)	Normalizes a nonterminal contents.
Tree	normalizeWholeTree(Tree tree, TreeFactory tf)	Normalize a whole tree -- one can assume that this is the root.

Methods inherited from class edu.stanford.nlp.trees.BobChrisTreeNormalizer

cleanUpLabel, normalizeTerminal, transformTree

Methods inherited from class java.lang.Object

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

Methods inherited from interface java.util.function.Function

andThen, compose

Methods inherited from interface edu.stanford.nlp.trees.TreeTransformer

apply

Constructor Detail

ArabicTreeNormalizer

public ArabicTreeNormalizer(boolean retainNPTmp,
                            boolean markPRDverb,
                            boolean changeNoLabels,
                            boolean retainNPSbj,
                            boolean retainPPClr)

ArabicTreeNormalizer

public ArabicTreeNormalizer(boolean retainNPTmp,
                            boolean markPRDverb,
                            boolean changeNoLabels)

ArabicTreeNormalizer

public ArabicTreeNormalizer(boolean retainNPTmp,
                            boolean markPRDverb)

ArabicTreeNormalizer

public ArabicTreeNormalizer(boolean retainNPTmp)

ArabicTreeNormalizer

public ArabicTreeNormalizer()

Method Detail

normalizeNonterminal

public String normalizeNonterminal(String category)

Description copied from class: BobChrisTreeNormalizer

Normalizes a nonterminal contents. This implementation strips functional tags, etc. and interns the nonterminal.

Overrides:

normalizeNonterminal in class BobChrisTreeNormalizer

Parameters:

category - The String that decorates this nonterminal node

Returns:

The normalized form of this nonterminal String

normalizeWholeTree

public Tree normalizeWholeTree(Tree tree,
                               TreeFactory tf)

Description copied from class: BobChrisTreeNormalizer

Normalize a whole tree -- one can assume that this is the root. This implementation deletes empty elements (ones with nonterminal tag label '-NONE-') from the tree, and splices out unary A over A nodes. It assumes that it is not given a null tree, but it may return one if there are no real words.

Overrides:

normalizeWholeTree in class BobChrisTreeNormalizer

Parameters:

tree - The tree to be normalized

tf - the TreeFactory to create new nodes (if needed)

Returns:

The normalized tree. May be null which means to not use this tree at all.