Examples with Mention edu.stanford.nlp.coref.data.Mention used on opensource projects

Example 56 with Mention

use of edu.stanford.nlp.coref.data.Mention in project CoreNLP by stanfordnlp.

the class HybridCorefPrinter method linkDistanceAnalysis.

public static void linkDistanceAnalysis(String[] args) throws Exception {
    Properties props = StringUtils.argsToProperties(args);
    HybridCorefSystem cs = new HybridCorefSystem(props);
    cs.docMaker.resetDocs();
    Counter<Integer> proper = new ClassicCounter<>();
    Counter<Integer> common = new ClassicCounter<>();
    Counter<Integer> pronoun = new ClassicCounter<>();
    Counter<Integer> list = new ClassicCounter<>();
    while (true) {
        Document document = cs.docMaker.nextDoc();
        if (document == null)
            break;
        for (int sentIdx = 0; sentIdx < document.predictedMentions.size(); sentIdx++) {
            List<Mention> predictedInSent = document.predictedMentions.get(sentIdx);
            for (int mIdx = 0; mIdx < predictedInSent.size(); mIdx++) {
                Mention m = predictedInSent.get(mIdx);
                loop: for (int distance = 0; distance <= sentIdx; distance++) {
                    List<Mention> candidates = Sieve.getOrderedAntecedents(m, sentIdx - distance, mIdx, document.predictedMentions, cs.dictionaries);
                    for (Mention candidate : candidates) {
                        if (candidate == m)
                            continue;
                        // ignore cataphora
                        if (distance == 0 && m.appearEarlierThan(candidate))
                            continue;
                        if (candidate.goldCorefClusterID == m.goldCorefClusterID) {
                            switch(m.mentionType) {
                                case NOMINAL:
                                    if (candidate.mentionType == MentionType.NOMINAL || candidate.mentionType == MentionType.PROPER) {
                                        common.incrementCount(distance);
                                        break loop;
                                    }
                                    break;
                                case PROPER:
                                    if (candidate.mentionType == MentionType.PROPER) {
                                        proper.incrementCount(distance);
                                        break loop;
                                    }
                                    break;
                                case PRONOMINAL:
                                    pronoun.incrementCount(distance);
                                    break loop;
                                case LIST:
                                    if (candidate.mentionType == MentionType.LIST) {
                                        list.incrementCount(distance);
                                        break loop;
                                    }
                                    break;
                                default:
                                    break;
                            }
                        }
                    }
                }
            }
        }
    }
    System.out.println("PROPER -------------------------------------------");
    Counters.printCounterSortedByKeys(proper);
    System.out.println("COMMON -------------------------------------------");
    Counters.printCounterSortedByKeys(common);
    System.out.println("PRONOUN -------------------------------------------");
    Counters.printCounterSortedByKeys(pronoun);
    System.out.println("LIST -------------------------------------------");
    Counters.printCounterSortedByKeys(list);
    log.info();
}

Example 57 with Mention

use of edu.stanford.nlp.coref.data.Mention in project CoreNLP by stanfordnlp.

the class CorefPrinter method printConllOutput.

public static String printConllOutput(Document document, List<List<Mention>> orderedMentions, boolean gold) {
    Annotation anno = document.annotation;
    List<List<String[]>> conllDocSentences = document.conllDoc.sentenceWordLists;
    String docID = anno.get(CoreAnnotations.DocIDAnnotation.class);
    StringBuilder sb = new StringBuilder();
    sb.append("#begin document ").append(docID).append("\n");
    List<CoreMap> sentences = anno.get(CoreAnnotations.SentencesAnnotation.class);
    for (int sentNum = 0; sentNum < sentences.size(); sentNum++) {
        List<CoreLabel> sentence = sentences.get(sentNum).get(CoreAnnotations.TokensAnnotation.class);
        List<String[]> conllSentence = conllDocSentences.get(sentNum);
        Map<Integer, Set<Mention>> mentionBeginOnly = Generics.newHashMap();
        Map<Integer, Set<Mention>> mentionEndOnly = Generics.newHashMap();
        Map<Integer, Set<Mention>> mentionBeginEnd = Generics.newHashMap();
        for (int i = 0; i < sentence.size(); i++) {
            mentionBeginOnly.put(i, new LinkedHashSet<>());
            mentionEndOnly.put(i, new LinkedHashSet<>());
            mentionBeginEnd.put(i, new LinkedHashSet<>());
        }
        for (Mention m : orderedMentions.get(sentNum)) {
            if (m.startIndex == m.endIndex - 1) {
                mentionBeginEnd.get(m.startIndex).add(m);
            } else {
                mentionBeginOnly.get(m.startIndex).add(m);
                mentionEndOnly.get(m.endIndex - 1).add(m);
            }
        }
        for (int i = 0; i < sentence.size(); i++) {
            StringBuilder sb2 = new StringBuilder();
            for (Mention m : mentionBeginOnly.get(i)) {
                if (sb2.length() > 0) {
                    sb2.append("|");
                }
                int corefClusterId = (gold) ? m.goldCorefClusterID : m.corefClusterID;
                sb2.append("(").append(corefClusterId);
            }
            for (Mention m : mentionBeginEnd.get(i)) {
                if (sb2.length() > 0) {
                    sb2.append("|");
                }
                int corefClusterId = (gold) ? m.goldCorefClusterID : m.corefClusterID;
                sb2.append("(").append(corefClusterId).append(")");
            }
            for (Mention m : mentionEndOnly.get(i)) {
                if (sb2.length() > 0) {
                    sb2.append("|");
                }
                int corefClusterId = (gold) ? m.goldCorefClusterID : m.corefClusterID;
                sb2.append(corefClusterId).append(")");
            }
            if (sb2.length() == 0)
                sb2.append("-");
            String[] columns = conllSentence.get(i);
            for (int j = 0; j < columns.length - 1; j++) {
                String column = columns[j];
                sb.append(column).append("\t");
            }
            sb.append(sb2).append("\n");
        }
        sb.append("\n");
    }
    sb.append("#end document").append("\n");
    return sb.toString();
}

Example 58 with Mention

use of edu.stanford.nlp.coref.data.Mention in project CoreNLP by stanfordnlp.

the class CorefRules method getSpeakerClusterId.

/**
   * Given the name of a speaker, returns the coref cluster id it belongs to (-1 if no cluster)
   * @param document The document to search in
   * @param speakerString The name to search for
   * @return cluster id
   */
public static int getSpeakerClusterId(Document document, String speakerString) {
    int speakerClusterId = -1;
    // try looking up cluster id from speaker info
    SpeakerInfo speakerInfo = null;
    if (speakerString != null) {
        speakerInfo = document.getSpeakerInfo(speakerString);
        if (speakerInfo != null) {
            speakerClusterId = speakerInfo.getCorefClusterId();
        }
    }
    if (speakerClusterId < 0 && speakerString != null && NumberMatchingRegex.isDecimalInteger(speakerString)) {
        // speakerString is number so is mention id
        try {
            int speakerMentionId = Integer.parseInt(speakerString);
            Mention mention = document.predictedMentionsByID.get(speakerMentionId);
            if (mention != null) {
                speakerClusterId = mention.corefClusterID;
                if (speakerInfo != null)
                    speakerInfo.addMention(mention);
            }
        } catch (Exception e) {
        }
    }
    return speakerClusterId;
}

Example 59 with Mention

use of edu.stanford.nlp.coref.data.Mention in project CoreNLP by stanfordnlp.

the class CorefRules method entityAlias.

public static boolean entityAlias(CorefCluster mentionCluster, CorefCluster potentialAntecedent, Semantics semantics, Dictionaries dict) throws Exception {
    Mention mention = mentionCluster.getRepresentativeMention();
    Mention antecedent = potentialAntecedent.getRepresentativeMention();
    if (mention.mentionType != MentionType.PROPER || antecedent.mentionType != MentionType.PROPER)
        return false;
    Method meth = semantics.wordnet.getClass().getMethod("alias", new Class[] { Mention.class, Mention.class });
    if ((Boolean) meth.invoke(semantics.wordnet, new Object[] { mention, antecedent })) {
        return true;
    }
    return false;
}

Example 60 with Mention

use of edu.stanford.nlp.coref.data.Mention in project CoreNLP by stanfordnlp.

the class CorefUtils method mergeCoreferenceClusters.

public static void mergeCoreferenceClusters(Pair<Integer, Integer> mentionPair, Document document) {
    Mention m1 = document.predictedMentionsByID.get(mentionPair.first);
    Mention m2 = document.predictedMentionsByID.get(mentionPair.second);
    if (m1.corefClusterID == m2.corefClusterID) {
        return;
    }
    int removeId = m1.corefClusterID;
    CorefCluster c1 = document.corefClusters.get(m1.corefClusterID);
    CorefCluster c2 = document.corefClusters.get(m2.corefClusterID);
    CorefCluster.mergeClusters(c2, c1);
    document.corefClusters.remove(removeId);
}