Examples with CorefMention edu.stanford.nlp.dcoref.CorefChain.CorefMention used on opensource projects

Example 1 with CorefMention

use of edu.stanford.nlp.dcoref.CorefChain.CorefMention in project CoreNLP by stanfordnlp.

the class DeterministicCorefAnnotator method addObsoleteCoreferenceAnnotations.

// for backward compatibility with a few old things
// TODO: Aim to get rid of this entirely
private static void addObsoleteCoreferenceAnnotations(Annotation annotation, List<List<Mention>> orderedMentions, Map<Integer, CorefChain> result) {
    List<Pair<IntTuple, IntTuple>> links = SieveCoreferenceSystem.getLinks(result);
    if (VERBOSE) {
        System.err.printf("Found %d coreference links:\n", links.size());
        for (Pair<IntTuple, IntTuple> link : links) {
            System.err.printf("LINK (%d, %d) -> (%d, %d)\n", link.first.get(0), link.first.get(1), link.second.get(0), link.second.get(1));
        }
    }
    //
    // save the coref output as CorefGraphAnnotation
    //
    // cdm 2013: this block didn't seem to be doing anything needed....
    // List<List<CoreLabel>> sents = new ArrayList<List<CoreLabel>>();
    // for (CoreMap sentence: annotation.get(CoreAnnotations.SentencesAnnotation.class)) {
    //   List<CoreLabel> tokens = sentence.get(CoreAnnotations.TokensAnnotation.class);
    //   sents.add(tokens);
    // }
    // this graph is stored in CorefGraphAnnotation -- the raw links found by the coref system
    List<Pair<IntTuple, IntTuple>> graph = new ArrayList<>();
    for (Pair<IntTuple, IntTuple> link : links) {
        //
        // Note: all offsets in the graph start at 1 (not at 0!)
        //       we do this for consistency reasons, as indices for syntactic dependencies start at 1
        //
        int srcSent = link.first.get(0);
        int srcTok = orderedMentions.get(srcSent - 1).get(link.first.get(1) - 1).headIndex + 1;
        int dstSent = link.second.get(0);
        int dstTok = orderedMentions.get(dstSent - 1).get(link.second.get(1) - 1).headIndex + 1;
        IntTuple dst = new IntTuple(2);
        dst.set(0, dstSent);
        dst.set(1, dstTok);
        IntTuple src = new IntTuple(2);
        src.set(0, srcSent);
        src.set(1, srcTok);
        graph.add(new Pair<>(src, dst));
    }
    annotation.set(CorefCoreAnnotations.CorefGraphAnnotation.class, graph);
    for (CorefChain corefChain : result.values()) {
        if (corefChain.getMentionsInTextualOrder().size() < 2)
            continue;
        Set<CoreLabel> coreferentTokens = Generics.newHashSet();
        for (CorefMention mention : corefChain.getMentionsInTextualOrder()) {
            CoreMap sentence = annotation.get(CoreAnnotations.SentencesAnnotation.class).get(mention.sentNum - 1);
            CoreLabel token = sentence.get(CoreAnnotations.TokensAnnotation.class).get(mention.headIndex - 1);
            coreferentTokens.add(token);
        }
        for (CoreLabel token : coreferentTokens) {
            token.set(CorefCoreAnnotations.CorefClusterAnnotation.class, coreferentTokens);
        }
    }
}

Example 2 with CorefMention

use of edu.stanford.nlp.dcoref.CorefChain.CorefMention in project CoreNLP by stanfordnlp.

the class SieveCoreferenceSystem method getLinks.

public static List<Pair<IntTuple, IntTuple>> getLinks(Map<Integer, CorefChain> result) {
    List<Pair<IntTuple, IntTuple>> links = new ArrayList<>();
    CorefChain.CorefMentionComparator comparator = new CorefChain.CorefMentionComparator();
    for (CorefChain c : result.values()) {
        List<CorefMention> s = c.getMentionsInTextualOrder();
        for (CorefMention m1 : s) {
            for (CorefMention m2 : s) {
                if (comparator.compare(m1, m2) == 1) {
                    links.add(new Pair<>(m1.position, m2.position));
                }
            }
        }
    }
    return links;
}

Example 3 with CorefMention

use of edu.stanford.nlp.dcoref.CorefChain.CorefMention in project CoreNLP by stanfordnlp.

the class SieveCoreferenceSystem method corefReturnHybridOutput.

public Map<Integer, edu.stanford.nlp.coref.data.CorefChain> corefReturnHybridOutput(Document document) throws Exception {
    // Multi-pass sieve coreference resolution
    for (int i = 0; i < sieves.length; i++) {
        currentSieve = i;
        DeterministicCorefSieve sieve = sieves[i];
        // Do coreference resolution using this pass
        coreference(document, sieve);
    }
    // post processing (e.g., removing singletons, appositions for conll)
    if ((!Constants.USE_GOLD_MENTIONS && doPostProcessing) || replicateCoNLL)
        postProcessing(document);
    // coref system output: edu.stanford.nlp.hcoref.data.CorefChain
    Map<Integer, edu.stanford.nlp.coref.data.CorefChain> result = Generics.newHashMap();
    for (CorefCluster c : document.corefClusters.values()) {
        // build mentionsMap and represents
        Map<IntPair, Set<edu.stanford.nlp.coref.data.CorefChain.CorefMention>> mentionsMap = Generics.newHashMap();
        IntPair keyPair = new IntPair(0, 0);
        mentionsMap.put(keyPair, new HashSet<>());
        Mention represents = null;
        edu.stanford.nlp.coref.data.CorefChain.CorefMention representsHybridVersion = null;
        for (Mention mention : c.getCorefMentions()) {
            // convert dcoref CorefMention to hcoref CorefMention
            //IntPair mentionPosition = new IntPair(mention.sentNum, mention.headIndex);
            IntTuple mentionPosition = document.positions.get(mention);
            CorefMention dcorefMention = new CorefMention(mention, mentionPosition);
            // tokens need the hcoref version of CorefClusterIdAnnotation
            mention.headWord.set(edu.stanford.nlp.coref.CorefCoreAnnotations.CorefClusterIdAnnotation.class, mention.corefClusterID);
            // drop the dcoref version of CorefClusterIdAnnotation
            mention.headWord.remove(CorefCoreAnnotations.CorefClusterIdAnnotation.class);
            // make the hcoref mention
            edu.stanford.nlp.coref.data.CorefChain.CorefMention hcorefMention = new edu.stanford.nlp.coref.data.CorefChain.CorefMention(edu.stanford.nlp.coref.data.Dictionaries.MentionType.valueOf(dcorefMention.mentionType.name()), edu.stanford.nlp.coref.data.Dictionaries.Number.valueOf(dcorefMention.number.name()), edu.stanford.nlp.coref.data.Dictionaries.Gender.valueOf(dcorefMention.gender.name()), edu.stanford.nlp.coref.data.Dictionaries.Animacy.valueOf(dcorefMention.animacy.name()), dcorefMention.startIndex, dcorefMention.endIndex, dcorefMention.headIndex, dcorefMention.corefClusterID, dcorefMention.mentionID, dcorefMention.sentNum, dcorefMention.position, dcorefMention.mentionSpan);
            mentionsMap.get(keyPair).add(hcorefMention);
            if (mention.moreRepresentativeThan(represents)) {
                represents = mention;
                representsHybridVersion = hcorefMention;
            }
        }
        edu.stanford.nlp.coref.data.CorefChain hybridCorefChain = new edu.stanford.nlp.coref.data.CorefChain(c.clusterID, mentionsMap, representsHybridVersion);
        result.put(c.clusterID, hybridCorefChain);
    }
    return result;
}