org.nlp2rdf.core
Class Text2RDF

java.lang.Object
  org.nlp2rdf.core.Text2RDF

public class Text2RDF
extends Object

Author:

Sebastian Hellmann

Field Summary

static String	stringOntologyUrl
static String	structuredSentenceOntologyUrl

Constructor Summary

Text2RDF()

Method Summary

void	addCopyOfOLiAClassesAndHierarchy(OLiAOntology olia, com.hp.hpl.jena.ontology.OntModel model)
void	addNextAndPreviousProperties(String prefix, String text, URIGenerator uriGenerator, com.hp.hpl.jena.ontology.OntModel model)
Document	createDocumentAnnotation(String prefix, String text, URIGenerator uriGenerator, com.hp.hpl.jena.ontology.OntModel model)
<S> S	createStringAnnotationForClass(Class<S> cl, String prefix, String text, Span span, URIGenerator uriGenerator, com.hp.hpl.jena.ontology.OntModel model) This is a convenience function, which does quite a lot: 1.
com.hp.hpl.jena.ontology.OntModel	generateNIFModel(String prefix, String text, TreeMap<Span,List<Span>> tokenizedText, URIGenerator uriGenerator, Document document, com.hp.hpl.jena.ontology.OntModel model) adds sso:Sentence and sso:Word to it adds the word property and additionally, though optional, the firstWord and lastWord property Note that the expected tokenizer matches the opennlp tokenizer well, so instead of forcing a tokenizer to match the interface it might be smarter to rewrite this method.
TreeMap<Span,List<Span>>	getTokenization(String prefix, String text, URIGenerator uriGenerator, com.hp.hpl.jena.ontology.OntModel model)

Methods inherited from class java.lang.Object

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

Field Detail

stringOntologyUrl

public static final String stringOntologyUrl

See Also:

Constant Field Values

structuredSentenceOntologyUrl

public static final String structuredSentenceOntologyUrl

See Also:

Constant Field Values

Constructor Detail

Text2RDF

public Text2RDF()

Method Detail

createDocumentAnnotation

public Document createDocumentAnnotation(String prefix,
                                         String text,
                                         URIGenerator uriGenerator,
                                         com.hp.hpl.jena.ontology.OntModel model)

Parameters:

prefix -

text -

uriGenerator -

model -

Returns:

createStringAnnotationForClass

public <S> S createStringAnnotationForClass(Class<S> cl,
                                            String prefix,
                                            String text,
                                            Span span,
                                            URIGenerator uriGenerator,
                                            com.hp.hpl.jena.ontology.OntModel model)

This is a convenience function, which does quite a lot: 1. generates the uri 2. add the recipe class, i.e. OffsetBased or ContexthashBased 3. adds the class which is given in class (must mbe from owl2java) 4. adds the anchorOf annotation Note: all changes are also reflected in model Note: if something goes wrong this method catches all exceptions, logs it and then re throws it as a runtime exception

Type Parameters:

S -

Parameters:

cl -

prefix -

text - the whole (reference) text

span - the span for the annotation

uriGenerator -

model -

Returns:

the Jena Individual with the Type give in cl

generateNIFModel

public com.hp.hpl.jena.ontology.OntModel generateNIFModel(String prefix,
                                                          String text,
                                                          TreeMap<Span,List<Span>> tokenizedText,
                                                          URIGenerator uriGenerator,
                                                          Document document,
                                                          com.hp.hpl.jena.ontology.OntModel model)

adds sso:Sentence and sso:Word to it adds the word property and additionally, though optional, the firstWord and lastWord property

Note that the expected tokenizer matches the opennlp tokenizer well, so instead of forcing a tokenizer to match the interface it might be smarter to rewrite this method.

Parameters:

prefix -

text -

tokenizedText -

uriGenerator -

document - use null if you want to opt out

model -

Returns:

the model filled with NIF

getTokenization

public TreeMap<Span,List<Span>> getTokenization(String prefix,
                                                String text,
                                                URIGenerator uriGenerator,
                                                com.hp.hpl.jena.ontology.OntModel model)

addNextAndPreviousProperties

public void addNextAndPreviousProperties(String prefix,
                                         String text,
                                         URIGenerator uriGenerator,
                                         com.hp.hpl.jena.ontology.OntModel model)

addCopyOfOLiAClassesAndHierarchy

public void addCopyOfOLiAClassesAndHierarchy(OLiAOntology olia,
                                             com.hp.hpl.jena.ontology.OntModel model)