edu.stanford.nlp.ie.machinereading.structure
Class ExtractionObject

java.lang.Object
  edu.stanford.nlp.ie.machinereading.structure.ExtractionObject

All Implemented Interfaces:

java.io.Serializable

Direct Known Subclasses:

EntityMention, RelationMention

public class ExtractionObject
extends java.lang.Object
implements java.io.Serializable

Represents any object that can be extracted - entity, relation, event

Author:

Andrey Gusev, Mihai

See Also:

Serialized Form

Field Summary

protected CoreMap	attributeMap This stores any optional attributes of ExtractionObjects
protected Span	extentTokenSpan Maximal token span relevant for this object, e.g., the largest NP for an entity mention The offsets are relative to the sentence that contains this object
protected java.lang.String	objectId Unique identifier of the object in its document
protected CoreMap	sentence Sentence that contains this object This assumes that each extraction object is intra-sentential (true in ACE, Roth, BioNLP, MR)
protected java.lang.String	subType Subtype, if available, e.g., GPE.CITY
protected java.lang.String	type Type of this mention, e.g., GPE
protected Counter<java.lang.String>	typeProbabilities Probabilities associated with this object We report probability values for each possible type for this object

Constructor Summary

ExtractionObject(java.lang.String objectId, CoreMap sentence, Span span, java.lang.String type, java.lang.String subtype)

Method Summary

CoreMap	attributeMap()
static java.lang.String	concatenateTypes(java.lang.String t1, java.lang.String t2) Concatenates two types
boolean	equals(java.lang.Object other)
java.lang.String	getDocumentId()
Span	getExtent()
java.lang.String	getExtentString()
int	getExtentTokenEnd()
int	getExtentTokenStart()
java.lang.String	getFullValue() Always returns the text corresponding to the extent of this object, even when getValue is overridden by subclass.
java.lang.String	getObjectId()
CoreMap	getSentence()
static Span	getSpan(ExtractionObject... objs) Returns the smallest span that covers the extent of all these objects
java.lang.String	getSubType()
java.lang.String	getType()
Counter<java.lang.String>	getTypeProbabilities()
java.lang.String	getValue() Returns the text corresponding to the extent of this object
boolean	printableObject(double beam, java.lang.String nilLabel) Returns true if it's worth saving/printing this object This happens in two cases: 1.
void	setExtent(Span s)
void	setSentence(CoreMap sent)
void	setType(java.lang.String t)
void	setTypeProbabilities(Counter<java.lang.String> probs)
static void	sortByExtent(java.util.List<ExtractionObject> objects)

Methods inherited from class java.lang.Object

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

Field Detail

objectId

protected final java.lang.String objectId

Unique identifier of the object in its document

sentence

protected CoreMap sentence

Sentence that contains this object This assumes that each extraction object is intra-sentential (true in ACE, Roth, BioNLP, MR)

type

protected java.lang.String type

Type of this mention, e.g., GPE

subType

protected final java.lang.String subType

Subtype, if available, e.g., GPE.CITY

extentTokenSpan

protected Span extentTokenSpan

Maximal token span relevant for this object, e.g., the largest NP for an entity mention The offsets are relative to the sentence that contains this object

attributeMap

protected CoreMap attributeMap

This stores any optional attributes of ExtractionObjects

typeProbabilities

protected Counter<java.lang.String> typeProbabilities

Probabilities associated with this object We report probability values for each possible type for this object

Constructor Detail

ExtractionObject

public ExtractionObject(java.lang.String objectId,
                        CoreMap sentence,
                        Span span,
                        java.lang.String type,
                        java.lang.String subtype)

Method Detail

getObjectId

public java.lang.String getObjectId()

getDocumentId

public java.lang.String getDocumentId()

getSentence

public CoreMap getSentence()

setSentence

public void setSentence(CoreMap sent)

getExtentTokenStart

public int getExtentTokenStart()

getExtentTokenEnd

public int getExtentTokenEnd()

getExtent

public Span getExtent()

setExtent

public void setExtent(Span s)

getExtentString

public java.lang.String getExtentString()

getType

public java.lang.String getType()

getSubType

public java.lang.String getSubType()

equals

public boolean equals(java.lang.Object other)

Overrides:

equals in class java.lang.Object

sortByExtent

public static void sortByExtent(java.util.List<ExtractionObject> objects)

getSpan

public static Span getSpan(ExtractionObject... objs)

Returns the smallest span that covers the extent of all these objects

Parameters:

objs -

getValue

public java.lang.String getValue()

Returns the text corresponding to the extent of this object

getFullValue

public final java.lang.String getFullValue()

Always returns the text corresponding to the extent of this object, even when getValue is overridden by subclass.

setType

public void setType(java.lang.String t)

concatenateTypes

public static java.lang.String concatenateTypes(java.lang.String t1,
                                                java.lang.String t2)

Concatenates two types

Parameters:

t1 -

t2 -

attributeMap

public CoreMap attributeMap()

setTypeProbabilities

public void setTypeProbabilities(Counter<java.lang.String> probs)

getTypeProbabilities

public Counter<java.lang.String> getTypeProbabilities()

printableObject

public boolean printableObject(double beam,
                               java.lang.String nilLabel)

Returns true if it's worth saving/printing this object This happens in two cases: 1. The type of the object is not nilLabel 2. The type of the object is nilLabel but the second ranked label is within the given beam (0 -- 100) of the first choice

Parameters:

beam -

nilLabel -