edu.stanford.nlp.patterns

Class PatternFactory

java.lang.Object

edu.stanford.nlp.patterns.PatternFactory

Direct Known Subclasses:

DepPatternFactory, SurfacePatternFactory

public class PatternFactory
extends java.lang.Object

Created by sonalg on 10/27/14.

Nested Class Summary

Nested Classes

Modifier and Type	Class and Description
static class	PatternFactory.PatternType

Field Summary

Fields

Modifier and Type	Field and Description
static java.util.List<java.lang.String>	fillerWords
static java.util.regex.Pattern	ignoreWordRegex by default doesn't ignore anything.
static java.lang.String	numWordsCompound Can just write a number (if same for all labels) or "Label1,2;Label2,3;...."
static java.util.Map<java.lang.String,java.lang.Integer>	numWordsCompoundMapped
static int	numWordsCompoundMax
static boolean	useLemmaContextTokens Use lemma instead of words for the context tokens
static boolean	useNER
static boolean	useStopWordsBeforeTerm allow to match stop words before a target term.
static boolean	useTargetNERRestriction Add NER restriction to the target phrase in the patterns

Constructor Summary

Constructors

Constructor and Description
PatternFactory()

Method Summary

Modifier and Type	Method and Description
static boolean	doNotUse(java.lang.String word, java.util.Set<CandidatePhrase> stopWords)
static java.util.Map<java.lang.Integer,java.util.Set>	getPatternsAroundTokens(PatternFactory.PatternType patternType, DataInstance sent, java.util.Set<CandidatePhrase> stopWords)
static void	setUp(java.util.Properties props, PatternFactory.PatternType patternType, java.util.Set<java.lang.String> labels)

Methods inherited from class java.lang.Object

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

Field Detail

useStopWordsBeforeTerm

@ArgumentParser.Option(name="useStopWordsBeforeTerm")
public static boolean useStopWordsBeforeTerm

allow to match stop words before a target term. This is to match something like "I am on some X" if the pattern is "I am on X"

useTargetNERRestriction

@ArgumentParser.Option(name="useTargetNERRestriction")
public static boolean useTargetNERRestriction

Add NER restriction to the target phrase in the patterns

useNER

@ArgumentParser.Option(name="useNER")
public static boolean useNER

numWordsCompound

@ArgumentParser.Option(name="numWordsCompound")
public static java.lang.String numWordsCompound

Can just write a number (if same for all labels) or "Label1,2;Label2,3;...."

numWordsCompoundMapped

public static java.util.Map<java.lang.String,java.lang.Integer> numWordsCompoundMapped

numWordsCompoundMax

public static int numWordsCompoundMax

useLemmaContextTokens

@ArgumentParser.Option(name="useLemmaContextTokens")
public static boolean useLemmaContextTokens

Use lemma instead of words for the context tokens

fillerWords

public static java.util.List<java.lang.String> fillerWords

ignoreWordRegex

public static java.util.regex.Pattern ignoreWordRegex

by default doesn't ignore anything. What phrases to ignore.

Constructor Detail

PatternFactory

public PatternFactory()

Method Detail

setUp

public static void setUp(java.util.Properties props,
                         PatternFactory.PatternType patternType,
                         java.util.Set<java.lang.String> labels)

doNotUse

public static boolean doNotUse(java.lang.String word,
                               java.util.Set<CandidatePhrase> stopWords)

getPatternsAroundTokens

public static java.util.Map<java.lang.Integer,java.util.Set> getPatternsAroundTokens(PatternFactory.PatternType patternType,
                                                                                     DataInstance sent,
                                                                                     java.util.Set<CandidatePhrase> stopWords)