Examples with RDFTerm org.apache.clerezza.commons.rdf.RDFTerm used on opensource projects

Example 56 with RDFTerm

use of org.apache.clerezza.commons.rdf.RDFTerm in project stanbol by apache.

the class CeliLemmatizerEnhancementEngineTest method validateMorphoFeatureProperty.

/**
     * [1..*] values of an {@link TypedLiteral} in the form {key=value}
     * @param enhancements The graph with the enhancements
     * @param textAnnotation the TextAnnotation to check
     */
private void validateMorphoFeatureProperty(Graph enhancements, BlankNodeOrIRI textAnnotation) {
    //This taste checks for known morpho features of a given input (constant TERM)
    Iterator<Triple> morphoFeatureIterator = enhancements.filter(textAnnotation, RDF_TYPE, null);
    assertTrue("No POS Morpho Feature value found for TextAnnotation " + textAnnotation + "!", morphoFeatureIterator.hasNext());
    while (morphoFeatureIterator.hasNext()) {
        RDFTerm morphoFeature = morphoFeatureIterator.next().getObject();
        assertTrue("Morpho Feature value are expected of typed literal", morphoFeature instanceof IRI);
        String feature = ((IRI) morphoFeature).getUnicodeString();
        assertFalse("Morpho Feature MUST NOT be empty", feature.isEmpty());
        if (feature.startsWith(OLIA_NAMESPACE)) {
            String key = feature.substring(OLIA_NAMESPACE.length());
            LexicalCategory cat = LexicalCategory.valueOf(key);
            assertTrue("Part of Speech of " + TERM + " should be " + LexicalCategory.Noun, (cat == LexicalCategory.Noun));
        }
    }
    morphoFeatureIterator = enhancements.filter(textAnnotation, CeliMorphoFeatures.HAS_GENDER, null);
    assertTrue("No Gender Morpho Feature value found for TextAnnotation " + textAnnotation + "!", morphoFeatureIterator.hasNext());
    if (morphoFeatureIterator.hasNext()) {
        RDFTerm morphoFeature = morphoFeatureIterator.next().getObject();
        assertTrue("Morpho Feature value are expected of typed literal", morphoFeature instanceof IRI);
        String feature = ((IRI) morphoFeature).getUnicodeString();
        assertFalse("Morpho Feature MUST NOT be empty", feature.isEmpty());
        if (feature.startsWith(OLIA_NAMESPACE)) {
            String key = feature.substring(OLIA_NAMESPACE.length());
            Gender cat = Gender.valueOf(key);
            assertTrue("Gender of " + TERM + " should be " + Gender.Feminine, (cat == Gender.Feminine));
        }
    }
    morphoFeatureIterator = enhancements.filter(textAnnotation, CeliMorphoFeatures.HAS_NUMBER, null);
    assertTrue("No Number Morpho Feature value found for TextAnnotation " + textAnnotation + "!", morphoFeatureIterator.hasNext());
    if (morphoFeatureIterator.hasNext()) {
        RDFTerm morphoFeature = morphoFeatureIterator.next().getObject();
        assertTrue("Morpho Feature value are expected of typed literal", morphoFeature instanceof IRI);
        String feature = ((IRI) morphoFeature).getUnicodeString();
        assertFalse("Morpho Feature MUST NOT be empty", feature.isEmpty());
        if (feature.startsWith(OLIA_NAMESPACE)) {
            String key = feature.substring(OLIA_NAMESPACE.length());
            NumberFeature cat = NumberFeature.valueOf(key);
            assertTrue("Number of " + TERM + " should be " + Gender.Feminine, (cat == NumberFeature.Singular));
        }
    }
    morphoFeatureIterator = enhancements.filter(textAnnotation, CeliLemmatizerEnhancementEngine.hasLemmaForm, null);
    assertTrue("No Number Morpho Feature value found for TextAnnotation " + textAnnotation + "!", morphoFeatureIterator.hasNext());
    if (morphoFeatureIterator.hasNext()) {
        RDFTerm morphoFeature = morphoFeatureIterator.next().getObject();
        assertTrue("Lemma Forms value are expected of type Literal", morphoFeature instanceof Literal);
        assertFalse("Lemma forms MUST NOT be empty", ((Literal) morphoFeature).getLexicalForm().isEmpty());
        String feature = ((Literal) morphoFeature).getLexicalForm();
        assertTrue("Lemma of " + TERM + " should be " + TERM, (feature.equals(TERM)));
    }
}

Example 57 with RDFTerm

use of org.apache.clerezza.commons.rdf.RDFTerm in project stanbol by apache.

the class CeliSentimentAnalysisEngineTest method testInput.

private void testInput(String txt, String lang) throws EngineException, IOException {
    ContentItem ci = wrapAsContentItem(txt);
    try {
        // add a simple triple to statically define the language of the test content
        ci.getMetadata().add(new TripleImpl(ci.getUri(), DC_LANGUAGE, new PlainLiteralImpl(lang)));
        sentimentAnalysisEngine.computeEnhancements(ci);
        TestUtils.logEnhancements(ci);
        HashMap<IRI, RDFTerm> expectedValues = new HashMap<IRI, RDFTerm>();
        expectedValues.put(Properties.ENHANCER_EXTRACTED_FROM, ci.getUri());
        expectedValues.put(Properties.DC_CREATOR, LiteralFactory.getInstance().createTypedLiteral(sentimentAnalysisEngine.getClass().getName()));
        expectedValues.put(DC_TYPE, CeliConstants.SENTIMENT_EXPRESSION);
        int textAnnoNum = validateAllTextAnnotations(ci.getMetadata(), txt, expectedValues);
        log.info(textAnnoNum + " TextAnnotations found ...");
        assertTrue("2 sentiment expressions should be recognized in: " + txt, textAnnoNum == 2);
        int entityAnnoNum = EnhancementStructureHelper.validateAllEntityAnnotations(ci.getMetadata(), expectedValues);
        assertTrue("0 entity annotations should be recognized in: " + txt, entityAnnoNum == 0);
    } catch (EngineException e) {
        RemoteServiceHelper.checkServiceUnavailable(e);
    }
}

Example 58 with RDFTerm

use of org.apache.clerezza.commons.rdf.RDFTerm in project stanbol by apache.

the class CeliClassificationEnhancementEngineTest method tesetEngine.

@Test
public void tesetEngine() throws Exception {
    ContentItem ci = wrapAsContentItem(TEXT);
    try {
        //add a simple triple to statically define the language of the test
        //content
        ci.getMetadata().add(new TripleImpl(ci.getUri(), DC_LANGUAGE, new PlainLiteralImpl("fr")));
        //unit test should not depend on each other (if possible)
        //CeliLanguageIdentifierEnhancementEngineTest.addEnanchements(ci);
        classificationEngine.computeEnhancements(ci);
        TestUtils.logEnhancements(ci);
        HashMap<IRI, RDFTerm> expectedValues = new HashMap<IRI, RDFTerm>();
        expectedValues.put(Properties.ENHANCER_EXTRACTED_FROM, ci.getUri());
        expectedValues.put(Properties.DC_CREATOR, LiteralFactory.getInstance().createTypedLiteral(classificationEngine.getClass().getName()));
        int textAnnoNum = EnhancementStructureHelper.validateAllTextAnnotations(ci.getMetadata(), TEXT, expectedValues);
        assertEquals("Only a single fise:TextAnnotation is expeted", 1, textAnnoNum);
        int numTopicAnnotations = validateAllTopicAnnotations(ci.getMetadata(), expectedValues);
        assertTrue("No TpocisAnnotations found", numTopicAnnotations > 0);
    } catch (EngineException e) {
        RemoteServiceHelper.checkServiceUnavailable(e);
        return;
    }
}

Example 59 with RDFTerm

use of org.apache.clerezza.commons.rdf.RDFTerm in project stanbol by apache.

the class CeliLanguageIdentifierEnhancementEngineTest method tesetEngine.

@Test
public void tesetEngine() throws Exception {
    ContentItem ci = wrapAsContentItem(TEXT);
    try {
        langIdentifier.computeEnhancements(ci);
        TestUtils.logEnhancements(ci);
        HashMap<IRI, RDFTerm> expectedValues = new HashMap<IRI, RDFTerm>();
        expectedValues.put(Properties.ENHANCER_EXTRACTED_FROM, ci.getUri());
        expectedValues.put(Properties.DC_CREATOR, LiteralFactory.getInstance().createTypedLiteral(langIdentifier.getClass().getName()));
        int numTextAnnotations = validateAllTextAnnotations(ci.getMetadata(), TEXT, expectedValues);
        assertEquals("A single TextAnnotation is expected by this Test", 1, numTextAnnotations);
        //even through this tests do not validate service quality but rather
        //the correct integration of the CELI service as EnhancementEngine
        //we expect the "fr" is detected for the parsed text
        assertEquals("The detected language for text '" + TEXT + "' MUST BE 'fr'", "fr", EnhancementEngineHelper.getLanguage(ci));
        int entityAnnoNum = validateAllEntityAnnotations(ci.getMetadata(), expectedValues);
        assertEquals("No EntityAnnotations are expected", 0, entityAnnoNum);
    } catch (EngineException e) {
        RemoteServiceHelper.checkServiceUnavailable(e);
    }
}

Example 60 with RDFTerm

use of org.apache.clerezza.commons.rdf.RDFTerm in project stanbol by apache.

the class CeliLemmatizerEnhancementEngineTest method validateLemmaFormProperty.

/**
     * [1..*] values of an {@link PlainLiteral} in the same language as the
     * analyzed text
     * @param enhancements The graph with the enhancements
     * @param textAnnotation the TextAnnotation to check
     * @param lang the language of the analyzed text
     * @return The number of lemma forms found
     */
private int validateLemmaFormProperty(Graph enhancements, BlankNodeOrIRI textAnnotation, String lang) {
    Iterator<Triple> lemmaFormsIterator = enhancements.filter(textAnnotation, hasLemmaForm, null);
    assertTrue("No lemma form value found for TextAnnotation " + textAnnotation + "!", lemmaFormsIterator.hasNext());
    int lemmaFormCount = 0;
    while (lemmaFormsIterator.hasNext()) {
        lemmaFormCount++;
        RDFTerm lemmaForms = lemmaFormsIterator.next().getObject();
        assertTrue("Lemma Forms value are expected of type Literal", lemmaForms instanceof Literal);
        assertFalse("Lemma forms MUST NOT be empty", ((Literal) lemmaForms).getLexicalForm().isEmpty());
        assertNotNull("Language of the Lemma Form literal MUST BE not null", ((Literal) lemmaForms).getLanguage());
        assertEquals("Language of the Lemma Form literal MUST BE the same as for the parsed text", lang, ((Literal) lemmaForms).getLanguage().toString());
    }
    return lemmaFormCount;
}