org.corpus_tools.salt.samples

Class SampleGenerator

java.lang.Object

org.corpus_tools.salt.samples.SampleGenerator

public class SampleGenerator
extends Object

Creates samples of SDocumentGraph and SCorpusGraph instances.

Author:

Florian Zipser

Field Summary

Fields

Modifier and Type	Field and Description
static String	LANG_DE iso 639-1 language code for german
static String	LANG_EN iso 639-1 language code for english
static String	MORPHOLOGY_LAYER The name of the morphologic layer containing the tokens.
static String	PRIMARY_TEXT_DE The primary text, which is used for the samples.
static String	PRIMARY_TEXT_EN The primary text, which is used for the samples.
static String	PRIMARY_TEXT_EN_SPK1 Primary text of speaker1
static String	PRIMARY_TEXT_EN_SPK2 Primary text of speaker2
static String	SYNTAX_LAYER

Constructor Summary

Constructors

Constructor and Description
SampleGenerator()

Method Summary

Modifier and Type	Method and Description
static void	createAnaphoricAnnotations(SDocument document)
static SCorpusGraph	createCorpusStructure_simple() Creates the following structure: rootCorpus | doc1
static SCorpusGraph	createCorpusStructure() Creates the following structure: rootCorpus / \ subCorpus1 subCorpus2 / \ / \ doc1 doc2 doc3 doc4
static SCorpusGraph	createCorpusStructure(SaltProject saltProject) Creates the following corpus structure and adds it to the given salt project.
static SCorpusGraph	createCorpusStructure(SCorpusGraph corpGraph1) Creates the following structure: rootCorpus / \ subCorpus1 subCorpus2 / \ / \ doc1 doc2 doc3 doc4
static void	createDependencies(SDocument document) This method creates the sample's dependency annotation.
static void	createDialogue(SDocument document) Creates a SDocumentGraph containing to texts of two different speakers, who are aligned via the STimeline related to the SToken objects.
static void	createDocumentStructure(SDocument document) Creates a document structure containing: primary text tokenization morphological annotations information structure annotation syntactical annotation anaphoric annotation
static void	createInformationStructureAnnotations(SDocument document) Annotates the SSpan objects above the tokenization with information structural annotations.
static void	createInformationStructureSpan(SDocument document) Creates SSpan object above the tokenization.
static void	createMorphologyAnnotations(SDocument document) Creates morphological annotations (pos and lemma) for the tokenized sample and adds them to each SToken object as SPOSAnnotation or SLemmaAnnotation object.
static void	createParallelData(SDocument document)
static void	createParallelData(SDocument document, boolean setTypeForPointRel) Creates a small parallel corpus, containing an english and a german text.
static STextualDS	createPrimaryData(SDocument document) Creates an STextualDS object containing the primary text PRIMARY_TEXT_EN and adds the object to the SDocumentGraph being contained by the given SDocument object.
static STextualDS	createPrimaryData(SDocument document, String language) Creates a STextualDS object containing the primary text PRIMARY_TEXT_EN, which is either an english text or its german translation and adds the object to the SDocumentGraph being contained by the given SDocument object.
static SaltProject	createSaltProject() Creates a complete SaltProject object having the complex structure rootCorpus / \ subCorpus1 subCorpus2 / \ / \ doc1 doc2 doc3 doc4
static void	createSyntaxAnnotations(SDocument document) This method creates the categorical annotations for the nodes of the sample syntax tree created in createSyntaxStructure(SDocument).
static void	createSyntaxStructure(SDocument document) Creates a syntax structure for the given SDocument object.
static SToken	createToken(int start, int end, STextualDS textualDS, SDocument document, SLayer layer) Creates a SToken covering the passed position and returns it.
static void	createTokens(SDocument document) Creates a set of SToken objects tokenizing the primary text PRIMARY_TEXT_EN in to the following tokens: Is this example more complicated than it appears to be ?
static List<SToken>	createTokens(SDocument document, STextualDS textualDS) Creates a set of SToken objects tokenizing the primary text PRIMARY_TEXT_EN or PRIMARY_TEXT_DE depending on the given STextualDS object in to the following tokens: Is this example more complicated than it appears to be ?
static void	createUntypedParallelData(SDocument document) Creates a small parallel corpus, containing an english and a german text.

Methods inherited from class java.lang.Object

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

Field Detail

PRIMARY_TEXT_EN

public static final String PRIMARY_TEXT_EN

The primary text, which is used for the samples.

See Also:

Constant Field Values

PRIMARY_TEXT_EN_SPK1

public static final String PRIMARY_TEXT_EN_SPK1

Primary text of speaker1

See Also:

Constant Field Values

PRIMARY_TEXT_EN_SPK2

public static final String PRIMARY_TEXT_EN_SPK2

Primary text of speaker2

See Also:

Constant Field Values

PRIMARY_TEXT_DE

public static final String PRIMARY_TEXT_DE

The primary text, which is used for the samples.

See Also:

Constant Field Values

MORPHOLOGY_LAYER

public static final String MORPHOLOGY_LAYER

The name of the morphologic layer containing the tokens.

See Also:

Constant Field Values

LANG_EN

public static final String LANG_EN

iso 639-1 language code for english

See Also:

Constant Field Values

LANG_DE

public static final String LANG_DE

iso 639-1 language code for german

See Also:

Constant Field Values

SYNTAX_LAYER

public static final String SYNTAX_LAYER

See Also:

Constant Field Values

Constructor Detail

SampleGenerator

public SampleGenerator()

Method Detail

createCorpusStructure

public static SCorpusGraph createCorpusStructure(SaltProject saltProject)

Creates the following corpus structure and adds it to the given salt project.

 
            rootCorpus
           /         \
        subCorpus1           subCorpus2
        /      \            /       \
 doc1   doc2         doc3     doc4
 

Throws:

IOException

SAXException

createSaltProject

public static SaltProject createSaltProject()

Creates a complete SaltProject object having the complex structure

 
            rootCorpus
           /         \
        subCorpus1          subCorpus2
        /      \            /       \
 doc1   doc2         doc3     doc4
 

Returns:

createCorpusStructure

public static SCorpusGraph createCorpusStructure()

Creates the following structure:

            rootCorpus
       /                    \
  subCorpus1              subCorpus2
  /       \              /        \
 doc1     doc2         doc3      doc4
 

Throws:

IOException

SAXException

createCorpusStructure

public static SCorpusGraph createCorpusStructure(SCorpusGraph corpGraph1)

Creates the following structure:

             rootCorpus
       /                     \
  subCorpus1             subCorpus2
  /       \             /         \
 doc1    doc2         doc3       doc4
 

Throws:

IOException

SAXException

createCorpusStructure_simple

public static SCorpusGraph createCorpusStructure_simple()

Creates the following structure:

 rootCorpus | doc1
 

Throws:

IOException

SAXException

createDialogue

public static void createDialogue(SDocument document)

Creates a SDocumentGraph containing to texts of two different speakers, who are aligned via the STimeline related to the SToken objects. The texts are "Is this example more complicated than it appears to be?" and "Uhm oh yes!", which are tokenized by words. The words 'to' and 'Oh' have been said simultaneously and are overlapping via the timeline.

Parameters:

document - document to be filled

createPrimaryData

public static STextualDS createPrimaryData(SDocument document)

Creates an STextualDS object containing the primary text PRIMARY_TEXT_EN and adds the object to the SDocumentGraph being contained by the given SDocument object. TODO WHAT HAS BEEN SUPPOSED TO BE SHOWN HERE? THE ORIGINAL TEXT OR THE LINK TO THE STRING OBJECT?

Parameters:

document - the document, to which the created STextualDS object will be added FEHLT IN SAMPLE GENERATOR

Returns:

returns the created primary text

createPrimaryData

public static STextualDS createPrimaryData(SDocument document,
                                           String language)

Creates a STextualDS object containing the primary text PRIMARY_TEXT_EN, which is either an english text or its german translation and adds the object to the SDocumentGraph being contained by the given SDocument object.

Parameters:

document - the document, to which the created STextualDS object will be added

language - the language of the resource to be created, LANG_EN for english, LANG_DE for german

Returns:

returns the created STextualDS object

createTokens

public static void createTokens(SDocument document)

Creates a set of SToken objects tokenizing the primary text PRIMARY_TEXT_EN in to the following tokens:

• Is
• this
• example
• more
• complicated
• than
• it
• appears
• to
• be
• ?

The created SToken objects and corresponding STextualRelation objects are added to the given SDocument object.

Parameters:

document - the document, to which the created SToken objects will be added

createTokens

public static List<SToken> createTokens(SDocument document,
                                        STextualDS textualDS)

Creates a set of SToken objects tokenizing the primary text PRIMARY_TEXT_EN or PRIMARY_TEXT_DE depending on the given STextualDS object in to the following tokens:

• Is
• this
• example
• more
• complicated
• than
• it
• appears
• to
• be
• ?

or

• Ist
• dieses
• Beispiel
• komplizierter
• als
• es
• zu
• sein
• scheint
• ?

The created SToken objects and corresponding STextualRelation objects are added to the given SDocument object.

Parameters:

document - the document, to which the created SToken objects will be added

Returns:

list of created SToken objects

createToken

public static SToken createToken(int start,
                                 int end,
                                 STextualDS textualDS,
                                 SDocument document,
                                 SLayer layer)

Creates a SToken covering the passed position and returns it.

Parameters:

start -

end -

textualDS -

document -

layer -

Returns:

created SToken object

createParallelData

public static void createParallelData(SDocument document)

createParallelData

public static void createParallelData(SDocument document,
                                      boolean setTypeForPointRel)

Creates a small parallel corpus, containing an english and a german text. The english text is "Is this example more complicated than it appears to be?", the german text is "Ist dieses Beispiel komplizierter als es zu sein scheint?". Both are tokenized by word borders.

Parameters:

document - he document containing the STextualDS objects

createUntypedParallelData

public static void createUntypedParallelData(SDocument document)

Creates a small parallel corpus, containing an english and a german text. The english text is "Is this example more complicated than it appears to be?", the german text is "Ist dieses Beispiel komplizierter als es zu sein scheint?". Both are tokenized by word borders. TODO INSUFFICIENT DOCUMENTATION

Parameters:

document - he document containing the STextualDS objects

createMorphologyAnnotations

public static void createMorphologyAnnotations(SDocument document)

Creates morphological annotations (pos and lemma) for the tokenized sample and adds them to each SToken object as SPOSAnnotation or SLemmaAnnotation object.

token	pos	lemma
Is	VBZ	be
this	DT	this
example	NN	example
more	ABR	more
complicated	JJ	complicated
than	IN	than
it	PRP	it
appears	VBZ	appear
to	TO	to
be	VB	be

Parameters:

document - the document containing the SToken and STextualDS objects

createInformationStructureSpan

public static void createInformationStructureSpan(SDocument document)

Creates SSpan object above the tokenization.

contrast-focus	topic
Is	this	example	more	complicated	than	it	appears	to	be

Parameters:

document -

createInformationStructureAnnotations

public static void createInformationStructureAnnotations(SDocument document)

Annotates the SSpan objects above the tokenization with information structural annotations.

Parameters:

document -

createSyntaxStructure

public static void createSyntaxStructure(SDocument document)

Creates a syntax structure for the given SDocument object. If it does not already contain a primary text and a tokenization, this method calls createPrimaryData(SDocument) and createTokens(SDocument).

Parameters:

document -

createSyntaxAnnotations

public static void createSyntaxAnnotations(SDocument document)

This method creates the categorical annotations for the nodes of the sample syntax tree created in createSyntaxStructure(SDocument).

Parameters:

document -

createDependencies

public static void createDependencies(SDocument document)

This method creates the sample's dependency annotation.

Parameters:

document -

createAnaphoricAnnotations

public static void createAnaphoricAnnotations(SDocument document)

Parameters:

document -

createDocumentStructure

public static void createDocumentStructure(SDocument document)

Creates a document structure containing:

• primary text
• tokenization
• morphological annotations
• information structure annotation
• syntactical annotation
• anaphoric annotation

Parameters:

document -