Examples with ACEReaderWithTrueCaseFixer edu.illinois.cs.cogcomp.nlp.corpusreaders.ACEReaderWithTrueCaseFixer used on opensource projects

Example 1 with ACEReaderWithTrueCaseFixer

use of edu.illinois.cs.cogcomp.nlp.corpusreaders.ACEReaderWithTrueCaseFixer in project cogcomp-nlp by CogComp.

the class BIOTester method statistics.

public static void statistics() {
    int ace_nam = 0;
    int ace_nom = 0;
    int ace_pro = 0;
    int ere_nam = 0;
    int ere_nom = 0;
    int ere_pro = 0;
    int tac_nam = 0;
    int tac_nom = 0;
    try {
        ACEReaderWithTrueCaseFixer aceReader = new ACEReaderWithTrueCaseFixer("data/all", false);
        for (TextAnnotation ta : aceReader) {
            for (Constituent c : ta.getView(ViewNames.MENTION_ACE)) {
                if (c.getAttribute("EntityMentionType").equals("NAM")) {
                    ace_nam++;
                }
                if (c.getAttribute("EntityMentionType").equals("NOM")) {
                    ace_nom++;
                }
                if (c.getAttribute("EntityMentionType").equals("PRO")) {
                    ace_pro++;
                }
            }
        }
        EREMentionRelationReader ereReader = new EREMentionRelationReader(EREDocumentReader.EreCorpus.ENR3, "data/ere/data", false);
        for (XmlTextAnnotation xta : ereReader) {
            TextAnnotation ta = xta.getTextAnnotation();
            for (Constituent c : ta.getView(ViewNames.MENTION_ERE)) {
                if (c.getAttribute("EntityMentionType").equals("NAM")) {
                    ere_nam++;
                }
                if (c.getAttribute("EntityMentionType").equals("NOM")) {
                    ere_nom++;
                }
                if (c.getAttribute("EntityMentionType").equals("PRO")) {
                    ere_pro++;
                }
            }
        }
        ColumnFormatReader columnFormatReader = new ColumnFormatReader("data/tac/2016.nam");
        for (TextAnnotation ta : columnFormatReader) {
            for (Constituent c : ta.getView("MENTIONS")) {
                tac_nam++;
            }
        }
        columnFormatReader = new ColumnFormatReader("data/tac/2016.nom");
        for (TextAnnotation ta : columnFormatReader) {
            for (Constituent c : ta.getView("MENTIONS")) {
                tac_nom++;
            }
        }
    } catch (Exception e) {
        e.printStackTrace();
    }
    System.out.println("ACE_NAM: " + ace_nam);
    System.out.println("ACE_NOM: " + ace_nom);
    System.out.println("ACE_PRO: " + ace_pro);
    System.out.println("ERE_NAM: " + ere_nam);
    System.out.println("ERE_NOM: " + ere_nom);
    System.out.println("ERE_PRO: " + ere_pro);
    System.out.println("TAC_NAM: " + tac_nam);
    System.out.println("TAC_NOM: " + tac_nom);
}

Example 2 with ACEReaderWithTrueCaseFixer

use of edu.illinois.cs.cogcomp.nlp.corpusreaders.ACEReaderWithTrueCaseFixer in project cogcomp-nlp by CogComp.

the class ExtentReader method getTextAnnotations.

public List<TextAnnotation> getTextAnnotations() throws InvalidPortException, InvalidEndpointException, IOException, JWNLException, DatastoreException {
    List<TextAnnotation> ret = new ArrayList<>();
    if (_corpus.equals("ACE")) {
        ACEReaderWithTrueCaseFixer aceReader = null;
        POSAnnotator posAnnotator = new POSAnnotator();
        try {
            aceReader = new ACEReaderWithTrueCaseFixer(_path, false);
            for (TextAnnotation ta : aceReader) {
                ta.addView(posAnnotator);
                ret.add(ta);
            }
        } catch (Exception e) {
            e.printStackTrace();
        }
    }
    if (_corpus.equals("ERE")) {
        EREMentionRelationReader ereMentionRelationReader = null;
        POSAnnotator posAnnotator = new POSAnnotator();
        try {
            ereMentionRelationReader = new EREMentionRelationReader(EREDocumentReader.EreCorpus.ENR3, _path, false);
            for (XmlTextAnnotation xta : ereMentionRelationReader) {
                TextAnnotation ta = xta.getTextAnnotation();
                ta.addView(posAnnotator);
                ret.add(ta);
            }
        } catch (Exception e) {
            e.printStackTrace();
        }
    }
    if (_corpus.startsWith("COMBINED")) {
        String realCorpus = _corpus.split("-")[1];
        String mode = _corpus.split("-")[2];
        int fold = Integer.parseInt(_corpus.split("-")[3]);
        BIOCombinedReader bioCombinedReader = new BIOCombinedReader(fold, realCorpus + "-" + mode, "ALL", true);
        for (Object ta = bioCombinedReader.next(); ta != null; ta = bioCombinedReader.next()) {
            ret.add((TextAnnotation) ta);
        }
    }
    return ret;
}

Example 3 with ACEReaderWithTrueCaseFixer

use of edu.illinois.cs.cogcomp.nlp.corpusreaders.ACEReaderWithTrueCaseFixer in project cogcomp-nlp by CogComp.

the class RelationExtractionTest method testAnnotator.

@Test
public void testAnnotator() {
    File modelDir = null;
    try {
        Datastore ds = new Datastore(new ResourceConfigurator().getDefaultConfig());
        modelDir = ds.getDirectory("org.cogcomp.re", "ACE_TEST_DOCS", 1.1, false);
    } catch (Exception e) {
        e.printStackTrace();
    }
    try {
        ACEReaderWithTrueCaseFixer aceReader = new ACEReaderWithTrueCaseFixer(modelDir.getAbsolutePath() + File.separator + "ACE_TEST_DOCS", false);
        POSAnnotator pos_annotator = new POSAnnotator();
        ChunkerAnnotator chunker = new ChunkerAnnotator(true);
        chunker.initialize(new ChunkerConfigurator().getDefaultConfig());
        Properties stanfordProps = new Properties();
        stanfordProps.put("annotators", "pos, parse");
        stanfordProps.put("parse.originalDependencies", true);
        stanfordProps.put("parse.maxlen", Stanford331Configurator.STFRD_MAX_SENTENCE_LENGTH);
        stanfordProps.put("parse.maxtime", Stanford331Configurator.STFRD_TIME_PER_SENTENCE);
        POSTaggerAnnotator posAnnotator = new POSTaggerAnnotator("pos", stanfordProps);
        ParserAnnotator parseAnnotator = new ParserAnnotator("parse", stanfordProps);
        StanfordDepHandler stanfordDepHandler = new StanfordDepHandler(posAnnotator, parseAnnotator);
        MentionAnnotator mentionAnnotator = new MentionAnnotator("ACE_TYPE");
        RelationAnnotator relationAnnotator = new RelationAnnotator();
        for (TextAnnotation ta : aceReader) {
            ta.addView(pos_annotator);
            chunker.addView(ta);
            stanfordDepHandler.addView(ta);
            mentionAnnotator.addView(ta);
            relationAnnotator.addView(ta);
            View mentionView = ta.getView(ViewNames.MENTION);
            assertTrue(mentionView.getConstituents().size() > 0);
            View relationView = ta.getView(ViewNames.RELATION);
            assertTrue(relationView.getRelations().size() > 0);
        }
    } catch (Exception e) {
        e.printStackTrace();
    }
}