Examples with Document edu.stanford.nlp.dcoref.Document used on opensource projects

Example 1 with Document

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

the class DeterministicCorefAnnotator method annotate.

@Override
public void annotate(Annotation annotation) {
    try {
        List<Tree> trees = new ArrayList<>();
        List<List<CoreLabel>> sentences = new ArrayList<>();
        // extract trees and sentence words
        // we are only supporting the new annotation standard for this Annotator!
        boolean hasSpeakerAnnotations = false;
        if (annotation.containsKey(CoreAnnotations.SentencesAnnotation.class)) {
            // int sentNum = 0;
            for (CoreMap sentence : annotation.get(CoreAnnotations.SentencesAnnotation.class)) {
                List<CoreLabel> tokens = sentence.get(CoreAnnotations.TokensAnnotation.class);
                sentences.add(tokens);
                Tree tree = sentence.get(TreeCoreAnnotations.TreeAnnotation.class);
                trees.add(tree);
                // locking here is crucial for correct threading!
                SemanticGraph dependencies = SemanticGraphFactory.makeFromTree(tree, Mode.COLLAPSED, Extras.NONE, null, true);
                sentence.set(SemanticGraphCoreAnnotations.AlternativeDependenciesAnnotation.class, dependencies);
                if (!hasSpeakerAnnotations) {
                    // check for speaker annotations
                    for (CoreLabel t : tokens) {
                        if (t.get(CoreAnnotations.SpeakerAnnotation.class) != null) {
                            hasSpeakerAnnotations = true;
                            break;
                        }
                    }
                }
                MentionExtractor.mergeLabels(tree, tokens);
                MentionExtractor.initializeUtterance(tokens);
            }
        } else {
            log.error("this coreference resolution system requires SentencesAnnotation!");
            return;
        }
        if (hasSpeakerAnnotations) {
            annotation.set(CoreAnnotations.UseMarkedDiscourseAnnotation.class, true);
        }
        // extract all possible mentions
        // this is created for each new annotation because it is not threadsafe
        RuleBasedCorefMentionFinder finder = new RuleBasedCorefMentionFinder(allowReparsing);
        List<List<Mention>> allUnprocessedMentions = finder.extractPredictedMentions(annotation, 0, corefSystem.dictionaries());
        // add the relevant info to mentions and order them for coref
        Document document = mentionExtractor.arrange(annotation, sentences, trees, allUnprocessedMentions);
        List<List<Mention>> orderedMentions = document.getOrderedMentions();
        if (VERBOSE) {
            for (int i = 0; i < orderedMentions.size(); i++) {
                System.err.printf("Mentions in sentence #%d:%n", i);
                for (int j = 0; j < orderedMentions.get(i).size(); j++) {
                    log.info("\tMention #" + j + ": " + orderedMentions.get(i).get(j).spanToString());
                }
            }
        }
        Map<Integer, edu.stanford.nlp.coref.data.CorefChain> result = corefSystem.corefReturnHybridOutput(document);
        annotation.set(edu.stanford.nlp.coref.CorefCoreAnnotations.CorefChainAnnotation.class, result);
        if (OLD_FORMAT) {
            Map<Integer, CorefChain> oldResult = corefSystem.coref(document);
            addObsoleteCoreferenceAnnotations(annotation, orderedMentions, oldResult);
        }
    } catch (RuntimeException e) {
        throw e;
    } catch (Exception e) {
        throw new RuntimeException(e);
    }
}