Examples of org.apache.clerezza.rdf.core.Language

org.apache.clerezza.rdf.core.Language
Represents a language as expressed by the RDF 4646 language tag @author reto

        if (at == null && text == null) {
            log.warn("NULL was parsed as AnalysedText AND Text for content item " 
                    + ci.getUri() + ". One of the two MUST BE present! -> call ignored");
            return;
        }
        final Language language;
        if(lang != null && !lang.isEmpty()){
            language = new Language(lang);
        } else {
            language = null;
        }
        if(log.isDebugEnabled()){
            log.debug("findNamedEntities model={},  language={}, text=",

View Full Code Here

        List<SentimentInfo> sentimentInfos = summarizeSentiments(at, spanTypes);
        String detectedLang = EnhancementEngineHelper.getLanguage(ci);
        ci.getLock().writeLock().lock();
        try {
            writeSentimentEnhancements(ci,sentimentInfos,at,
                detectedLang == null ? null : new Language(detectedLang));
        } finally {
            ci.getLock().writeLock().unlock();
        }
        
    }

View Full Code Here

      this.addLemmatizationEnhancement(ci, text, language, graph);
    }
  }


  private void addMorphoAnalysisEnhancement(ContentItem ci, String text, String language, MGraph g) throws EngineException {
    Language lang = new Language(language); // clerezza language for PlainLiterals
    List<LexicalEntry> terms;
    try {
      terms = this.client.performMorfologicalAnalysis(text, language);
    } catch (IOException e) {
      throw new EngineException("Error while calling the CELI Lemmatizer" + " service (configured URL: " + serviceURL + ")!", e);

View Full Code Here

      ci.getLock().writeLock().unlock();
    }
  }


  private void addLemmatizationEnhancement(ContentItem ci, String text, String language, MGraph g) throws EngineException {
    Language lang = new Language(language); // clerezza language for PlainLiterals
    String lemmatizedContents;
    try {
      lemmatizedContents = this.client.lemmatizeContents(text, language);
    } catch (IOException e) {
      throw new EngineException("Error while calling the CELI Lemmatizer" + " service (configured URL: " + serviceURL + ")!", e);

View Full Code Here

     * @param ci
     * @param linkedEntities
     * @param language
     */
    private void writeEnhancements(ContentItem ci, Collection<LinkedEntity> linkedEntities, String language) {
        Language languageObject = null;
        if(language != null && !language.isEmpty()){
            languageObject = new Language(language);
        }
        MGraph metadata = ci.getMetadata();
        for(LinkedEntity linkedEntity : linkedEntities){
            Collection<UriRef> textAnnotations = new ArrayList<UriRef>(linkedEntity.getOccurrences().size());
            //first create the TextAnnotations for the Occurrences
            for(Occurrence occurrence : linkedEntity.getOccurrences()){
                UriRef textAnnotation = EnhancementEngineHelper.createTextEnhancement(ci, this);
                textAnnotations.add(textAnnotation);
                metadata.add(new TripleImpl(textAnnotation, 
                    Properties.ENHANCER_START, 
                    literalFactory.createTypedLiteral(occurrence.getStart())));
                metadata.add(new TripleImpl(textAnnotation, 
                    Properties.ENHANCER_END, 
                    literalFactory.createTypedLiteral(occurrence.getEnd())));
                metadata.add(new TripleImpl(textAnnotation, 
                    Properties.ENHANCER_SELECTION_CONTEXT, 
                    new PlainLiteralImpl(occurrence.getContext(),languageObject)));
                metadata.add(new TripleImpl(textAnnotation, 
                    Properties.ENHANCER_SELECTED_TEXT, 
                    new PlainLiteralImpl(occurrence.getSelectedText(),languageObject)));
                metadata.add(new TripleImpl(textAnnotation, 
                    Properties.ENHANCER_CONFIDENCE, 
                    literalFactory.createTypedLiteral(linkedEntity.getScore())));
                for(UriRef dcType : linkedEntity.getTypes()){
                    metadata.add(new TripleImpl(
                        textAnnotation, Properties.DC_TYPE, dcType));
                }
            }
            //now the EntityAnnotations for the Suggestions
            for(Suggestion suggestion : linkedEntity.getSuggestions()){
                UriRef entityAnnotation = EnhancementEngineHelper.createEntityEnhancement(ci, this);
                //should we use the label used for the match, or search the
                //representation for the best label ... currently its the matched one
                Text label = suggestion.getBestLabel(linkerConfig.getNameField(),language);
                metadata.add(new TripleImpl(entityAnnotation, 
                    Properties.ENHANCER_ENTITY_LABEL, 
                    label.getLanguage() == null ?
                            new PlainLiteralImpl(label.getText()) :
                                new PlainLiteralImpl(label.getText(),
                                    new Language(label.getLanguage()))));
                metadata.add(new TripleImpl(entityAnnotation, 
                    Properties.ENHANCER_ENTITY_REFERENCE, 
                    new UriRef(suggestion.getRepresentation().getId())));
                Iterator<Reference> suggestionTypes = suggestion.getRepresentation().getReferences(linkerConfig.getTypeField());
                while(suggestionTypes.hasNext()){

View Full Code Here

    }
        String language = EnhancementEngineHelper.getLanguage(ci);
        if (language == null) {
            throw new IllegalStateException("Unable to extract Language for " + "ContentItem " + ci.getUri() + ": This is also checked in the canEnhance " + "method! -> This indicated an Bug in the implementation of the " + "EnhancementJobManager!");
        }
        Language lang = new Language(language); //used for the palin literals in TextAnnotations
    try {
      List<SentimentExpression> lista = this.client.extractSentimentExpressions(text, language);
      LiteralFactory literalFactory = LiteralFactory.getInstance();


      MGraph g = ci.getMetadata();

View Full Code Here

        } //else the matched label will be the best to use
        Literal literal;
        if (label.getLanguage() == null) {
            literal = new PlainLiteralImpl(label.getText());
        } else {
            literal = new PlainLiteralImpl(label.getText(), new Language(label.getLanguage()));
        }
        // Now create the entityAnnotation
        UriRef entityAnnotation = EnhancementEngineHelper.createEntityEnhancement(graph, engine,
            contentItemId);
        // first relate this entity annotation to the text annotation(s)

View Full Code Here

        }
        if (text == null) {
            log.warn("NULL was parsed as text for content item " + ci.getUri().getUnicodeString() + "! -> call ignored");
            return;
        }
        final Language language;
        if(lang != null && !lang.isEmpty()){
            language = new Language(lang);
        } else {
            language = null;
        }
        log.debug("findNamedEntities typeUri={}, type={}, text=", 
                new Object[]{ typeUri, typeLabel, StringUtils.abbreviate(text, 100) });

View Full Code Here

     * @param occs a Collection of entity information
     * @param ci the content item
     */
    public void createEnhancements(Collection<CalaisEntityOccurrence> occs, ContentItem ci) {
        LiteralFactory literalFactory = LiteralFactory.getInstance();
        final Language language; // used for plain literals representing parts fo the content
        String langString = getMetadataLanguage(ci.getMetadata(), null);
        if(langString != null && !langString.isEmpty()){
            language = new Language(langString);
        } else {
            language = null;
        }
        //TODO create TextEnhancement (form, start, end, type?) and EntityAnnotation (id, name, type)
        HashMap<Resource, UriRef> entityAnnotationMap = new HashMap<Resource, UriRef>();

View Full Code Here

            Iterator<Triple> abstracts = entityProperties.filter(uri, SUMMARY, null);
            while (abstracts.hasNext()) {
                Resource object = abstracts.next().getObject();
                if (object instanceof PlainLiteral) {
                    PlainLiteral abstract_ = (PlainLiteral) object;
                    if (new Language("en").equals(abstract_.getLanguage())) {
                        return abstract_.getLexicalForm();
                    }
                }
            }
            return "";

View Full Code Here

0 1 2 3 4 5 6 7 8

TOP

Related Classes of org.apache.clerezza.rdf.core.Language

com.github.jsonldjava.clerezza.ClerezzaTripleCallback

org.apache.clerezza.jaxrs.sparql.providers.ResultSetJsonMessageBodyWriter

org.apache.clerezza.jaxrs.sparql.providers.ResultSetXmlMessageBodyWriter

org.apache.clerezza.platform.concepts.core.ConceptManipulator

org.apache.clerezza.platform.language.LanguageDescription

org.apache.clerezza.rdf.core.impl.PlainLiteralImplTest

org.apache.clerezza.rdf.core.sparql.QueryParserTest

org.apache.clerezza.rdf.core.test.MGraphTest

org.apache.clerezza.rdf.jena.commons.Jena2TriaUtil

org.apache.clerezza.rdf.rdfjson.parser.RdfJsonParsingProvider

All source code are property of their respective owners. Java is a trademark of Sun Microsystems, Inc and owned by ORACLE Inc. Contact coftware#gmail.com.