Example usage for edu.stanford.nlp.trees TypedDependency gov

List of usage examples for edu.stanford.nlp.trees TypedDependency gov

  1. HOME
  2. Java
  3. edu.stanford.nlp
  4. edu.stanford.nlp.trees.*
  5. TypedDependency
  6. gov

Introduction

In this page you can find the example usage for edu.stanford.nlp.trees TypedDependency gov.

Prototype

IndexedWord gov

To view the source code for edu.stanford.nlp.trees TypedDependency gov.

Click Source Link

Usage

From source file:DependencyParse.java

License:Apache License

public static void main(String[] args) throws Exception {
    Properties props = StringUtils.argsToProperties(args);
    if (!props.containsKey("tokpath") || !props.containsKey("parentpath") || !props.containsKey("relpath")) {
        System.err.println(/*from  w  ww.j  a v  a 2 s. c o m*/
                "usage: java DependencyParse -tokenize - -tokpath <tokpath> -parentpath <parentpath> -relpath <relpath>");
        System.exit(1);
    }

    boolean tokenize = false;
    if (props.containsKey("tokenize")) {
        tokenize = true;
    }

    String tokPath = props.getProperty("tokpath");
    String parentPath = props.getProperty("parentpath");
    String relPath = props.getProperty("relpath");

    BufferedWriter tokWriter = new BufferedWriter(new FileWriter(tokPath));
    BufferedWriter parentWriter = new BufferedWriter(new FileWriter(parentPath));
    BufferedWriter relWriter = new BufferedWriter(new FileWriter(relPath));

    MaxentTagger tagger = new MaxentTagger(TAGGER_MODEL);
    DependencyParser parser = DependencyParser.loadFromModelFile(PARSER_MODEL);
    Scanner stdin = new Scanner(System.in);
    int count = 0;
    long start = System.currentTimeMillis();
    while (stdin.hasNextLine()) {
        String line = stdin.nextLine();
        List<HasWord> tokens = new ArrayList<>();
        if (tokenize) {
            PTBTokenizer<Word> tokenizer = new PTBTokenizer(new StringReader(line), new WordTokenFactory(), "");
            for (Word label; tokenizer.hasNext();) {
                tokens.add(tokenizer.next());
            }
        } else {
            for (String word : line.split(" ")) {
                tokens.add(new Word(word));
            }
        }

        List<TaggedWord> tagged = tagger.tagSentence(tokens);

        int len = tagged.size();
        Collection<TypedDependency> tdl = parser.predict(tagged).typedDependencies();
        int[] parents = new int[len];
        for (int i = 0; i < len; i++) {
            // if a node has a parent of -1 at the end of parsing, then the node
            // has no parent.
            parents[i] = -1;
        }

        String[] relns = new String[len];
        for (TypedDependency td : tdl) {
            // let root have index 0
            int child = td.dep().index();
            int parent = td.gov().index();
            relns[child - 1] = td.reln().toString();
            parents[child - 1] = parent;
        }

        // print tokens
        StringBuilder sb = new StringBuilder();
        for (int i = 0; i < len - 1; i++) {
            if (tokenize) {
                sb.append(PTBTokenizer.ptbToken2Text(tokens.get(i).word()));
            } else {
                sb.append(tokens.get(i).word());
            }
            sb.append(' ');
        }
        if (tokenize) {
            sb.append(PTBTokenizer.ptbToken2Text(tokens.get(len - 1).word()));
        } else {
            sb.append(tokens.get(len - 1).word());
        }
        sb.append('\n');
        tokWriter.write(sb.toString());

        // print parent pointers
        sb = new StringBuilder();
        for (int i = 0; i < len - 1; i++) {
            sb.append(parents[i]);
            sb.append(' ');
        }
        sb.append(parents[len - 1]);
        sb.append('\n');
        parentWriter.write(sb.toString());

        // print relations
        sb = new StringBuilder();
        for (int i = 0; i < len - 1; i++) {
            sb.append(relns[i]);
            sb.append(' ');
        }
        sb.append(relns[len - 1]);
        sb.append('\n');
        relWriter.write(sb.toString());

        count++;
        if (count % 1000 == 0) {
            double elapsed = (System.currentTimeMillis() - start) / 1000.0;
            System.err.printf("Parsed %d lines (%.2fs)\n", count, elapsed);
        }
    }

    long totalTimeMillis = System.currentTimeMillis() - start;
    System.err.printf("Done: %d lines in %.2fs (%.1fms per line)\n", count, totalTimeMillis / 1000.0,
            totalTimeMillis / (double) count);
    tokWriter.close();
    parentWriter.close();
    relWriter.close();
}

From source file:ConstituencyParse.java

License:Apache License

public int[] depTreeParents(Tree tree, List<HasWord> tokens) {
    GrammaticalStructure gs = gsf.newGrammaticalStructure(tree);
    Collection<TypedDependency> tdl = gs.typedDependencies();
    int len = tokens.size();
    int[] parents = new int[len];
    for (int i = 0; i < len; i++) {
        // if a node has a parent of -1 at the end of parsing, then the node
        // has no parent.
        parents[i] = -1;/*w w w  .  ja  v a  2 s  .c o m*/
    }

    for (TypedDependency td : tdl) {
        // let root have index 0
        int child = td.dep().index();
        int parent = td.gov().index();
        parents[child - 1] = parent;
    }

    return parents;
}

From source file:ca.ualberta.exemplar.core.ParserMalt.java

License:Open Source License

@Override
public List<CoreMap> parseText(String text) {

    List<CoreMap> sentences = null;

    try {// w  ww.  j  av  a2  s .  c o  m
        Annotation document = new Annotation(text);
        pipeline.annotate(document);

        sentences = document.get(SentencesAnnotation.class);
        for (CoreMap sentence : sentences) {

            List<CoreLabel> tokens = sentence.get(TokensAnnotation.class);
            String[] conllInput = sentenceToCoNLLInput(tokens);

            DependencyGraph graph = (DependencyGraph) maltParser.parse(conllInput);

            Result result = graphToCoNLL(graph);
            List<List<String>> conll = result.conll;
            int rootIndex = result.rootIndex;

            EnglishGrammaticalStructure gs = EnglishGrammaticalStructure.buildCoNNLXGrammaticStructure(conll);

            TreeGraphNode root = null;

            List<TypedDependency> deps = gs.typedDependenciesCCprocessed();

            // Add root dependency and ner annotations
            int size = deps.size();
            for (int i = 0; i < size; i++) {
                TypedDependency td = deps.get(i);
                if (td.gov().index() == rootIndex) {
                    root = td.gov();
                    deps.add(new TypedDependency(GrammaticalRelation.ROOT, td.gov(), td.gov()));
                }
                {
                    TreeGraphNode n = td.dep();
                    if (n.label().ner() == null) {
                        n.label().setNER(tokens.get(n.index() - 1).ner());
                        n.label().setBeginPosition(tokens.get(n.index() - 1).beginPosition());
                        n.label().setEndPosition(tokens.get(n.index() - 1).endPosition());
                        n.label().setLemma(tokens.get(n.index() - 1).lemma());
                    }
                }
                {
                    TreeGraphNode n = td.gov();
                    if (n.label().ner() == null) {
                        n.label().setNER(tokens.get(n.index() - 1).ner());
                        n.label().setBeginPosition(tokens.get(n.index() - 1).beginPosition());
                        n.label().setEndPosition(tokens.get(n.index() - 1).endPosition());
                        n.label().setLemma(tokens.get(n.index() - 1).lemma());
                    }
                }
            }
            if (root == null)
                continue;

            List<TreeGraphNode> roots = new ArrayList<TreeGraphNode>();
            roots.add(gs.root());

            SemanticGraph sg = new SemanticGraph(deps, roots);
            sentence.set(CollapsedCCProcessedDependenciesAnnotation.class, sg);

        }

    } catch (Exception e) {
        e.printStackTrace();
    }

    return sentences;
}

From source file:cc.vidr.parseviz.ParseViz.java

License:Open Source License

public static void printDependenciesDot(Tree tree, StringBuilder sb) {
    sb.append("digraph{\n");
    for (TypedDependency td : typedDependencies(tree)) {
        GrammaticalRelation reln = td.reln();
        TreeGraphNode gov = td.gov();
        TreeGraphNode dep = td.dep();/*from   w ww  .ja  v a2 s . c om*/
        sb.append("n").append(gov.index()).append("[label=\"").append(gov).append("\"];\n").append("n")
                .append(dep.index()).append("[label=\"").append(dep).append("\"];\n").append("n")
                .append(gov.index()).append("->n").append(dep.index()).append("[label=\"").append(reln)
                .append("\"];\n");
    }
    sb.append("}\n");
}

From source file:com.search.MySearchHandler.java

License:Apache License

private static String compress(TermTokenStream queryStream) {

    List<TypedDependency> tdl = getTyped(queryStream.toNLPString());

    ListIterator<TypedDependency> tdllist = tdl.listIterator();
    int count = 1;
    while (tdllist.hasNext()) {

        TypedDependency typd = tdllist.next();
        // Have to fix the multiple words in the sequence issue
        if ((typd.reln().toString().equals("nn") | typd.reln().toString().equals("amod")
                | (typd.reln().toString().equals("det") && typd.dep().index() != 1)
                | typd.reln().toString().equals("num") | typd.reln().toString().equals("number"))
                && ((typd.gov().index() - typd.dep().index()) == 1
                        || (typd.gov().index() - typd.dep().index()) == -1)) {
            if ((typd.gov().index() - typd.dep().index()) == -1) {
                queryStream.mergeWithNext(typd.gov().index() - count);
                count++;/*w  w  w  .  j a va2  s .co  m*/
            } else {
                queryStream.mergeWithNext(typd.dep().index() - count);
                count++;
            }
        }
    }

    if (found(getTyped(queryStream.toNLPString()))) {
        return compress(queryStream);
    }

    return queryStream.toNLPString();
}

From source file:com.search.MySearchHandler.java

License:Apache License

private static boolean found(List<TypedDependency> typed) {
    ListIterator<TypedDependency> tdllist = typed.listIterator();
    boolean iftrue = false;
    while (tdllist.hasNext()) {

        TypedDependency typd = tdllist.next();
        // Have to fix the multiple words in the sequence issue
        if ((typd.reln().toString().equals("nn") | typd.reln().toString().equals("amod")
                | (typd.reln().toString().equals("det") && typd.dep().index() != 1)
                | typd.reln().toString().equals("num") | typd.reln().toString().equals("number"))
                && ((typd.gov().index() - typd.dep().index()) == 1
                        || (typd.gov().index() - typd.dep().index()) == -1)) {
            iftrue = true;//from w w w . j  a  v a 2s. c om
        }

    }
    return iftrue;
}

From source file:coreferenceresolver.element.Token.java

/**
 * @param newTokenSentiment the sentimentOrientation to set
 *///from  w ww .  j a va2  s  . c  o  m
public void setSentimentOrientation(int newTokenSentiment, Collection<TypedDependency> typedDeps) {
    if ((newTokenSentiment == Util.POSITIVE || newTokenSentiment == Util.NEGATIVE)
            && (!this.POS.equals("IN"))) {
        for (TypedDependency typedDependency : typedDeps) {
            if (typedDependency.reln().toString().equals("neg") && typedDependency.gov().value().equals(word)) {
                newTokenSentiment = Util.reverseSentiment(newTokenSentiment);
            }
            if (typedDependency.reln().toString().equals("mark")
                    && (typedDependency.dep().value().toLowerCase().equals("although")
                            || typedDependency.dep().value().toLowerCase().equals("though")
                            || typedDependency.dep().value().toLowerCase().equals("while"))
                    && typedDependency.gov().value().equals(word)) {
                newTokenSentiment = Util.NEUTRAL;
                break;
            }
        }
    }
    this.sentimentOrientation = newTokenSentiment;
}

From source file:de.tudarmstadt.ukp.dkpro.core.stanfordnlp.StanfordDependencyConverter.java

License:Open Source License

protected void doCreateDependencyTags(JCas aJCas, TreebankLanguagePack aLP, Tree parseTree,
        List<Token> tokens) {
    GrammaticalStructure gs;// w w  w. jav  a 2  s  . c om
    try {
        gs = aLP.grammaticalStructureFactory(aLP.punctuationWordRejectFilter(), aLP.typedDependencyHeadFinder())
                .newGrammaticalStructure(parseTree);
    } catch (UnsupportedOperationException e) {
        // We already warned in the model provider if dependencies are not supported, so here
        // we just do nothing and skip the dependencies.
        return;
    }

    Collection<TypedDependency> dependencies = null;
    switch (mode) {
    case BASIC:
        dependencies = gs.typedDependencies(); // gs.typedDependencies(false);
        break;
    case NON_COLLAPSED:
        dependencies = gs.allTypedDependencies(); // gs.typedDependencies(true);
        break;
    case COLLAPSED_WITH_EXTRA:
        dependencies = gs.typedDependenciesCollapsed(true);
        break;
    case COLLAPSED:
        dependencies = gs.typedDependenciesCollapsed(false);
        break;
    case CC_PROPAGATED:
        dependencies = gs.typedDependenciesCCprocessed(true);
        break;
    case CC_PROPAGATED_NO_EXTRA:
        dependencies = gs.typedDependenciesCCprocessed(false);
        break;
    case TREE:
        dependencies = gs.typedDependenciesCollapsedTree();
        break;
    }

    for (TypedDependency currTypedDep : dependencies) {
        int govIndex = currTypedDep.gov().index();
        int depIndex = currTypedDep.dep().index();
        if (govIndex != 0) {
            // Stanford CoreNLP produces a dependency relation between a verb and ROOT-0 which
            // is not token at all!
            Token govToken = tokens.get(govIndex - 1);
            Token depToken = tokens.get(depIndex - 1);

            StanfordAnnotator.createDependencyAnnotation(aJCas, currTypedDep.reln(), govToken, depToken);
        }
    }
}

From source file:de.tudarmstadt.ukp.dkpro.core.stanfordnlp.StanfordParser.java

License:Open Source License

protected void doCreateDependencyTags(ParserGrammar aParser, StanfordAnnotator sfAnnotator, Tree parseTree,
        List<Token> tokens) {
    GrammaticalStructure gs;/*from www .  j a v a2s  .c  om*/
    try {
        TreebankLanguagePack tlp = aParser.getTLPParams().treebankLanguagePack();
        gs = tlp.grammaticalStructureFactory(tlp.punctuationWordRejectFilter(), tlp.typedDependencyHeadFinder())
                .newGrammaticalStructure(parseTree);
    } catch (UnsupportedOperationException e) {
        // We already warned in the model provider if dependencies are not supported, so here
        // we just do nothing and skip the dependencies.
        return;
    }

    Collection<TypedDependency> dependencies = null;
    switch (mode) {
    case BASIC:
        dependencies = gs.typedDependencies(); // gs.typedDependencies(false);
        break;
    case NON_COLLAPSED:
        dependencies = gs.allTypedDependencies(); // gs.typedDependencies(true);
        break;
    case COLLAPSED_WITH_EXTRA:
        dependencies = gs.typedDependenciesCollapsed(true);
        break;
    case COLLAPSED:
        dependencies = gs.typedDependenciesCollapsed(false);
        break;
    case CC_PROPAGATED:
        dependencies = gs.typedDependenciesCCprocessed(true);
        break;
    case CC_PROPAGATED_NO_EXTRA:
        dependencies = gs.typedDependenciesCCprocessed(false);
        break;
    case TREE:
        dependencies = gs.typedDependenciesCollapsedTree();
        break;
    }

    for (TypedDependency currTypedDep : dependencies) {
        int govIndex = currTypedDep.gov().index();
        int depIndex = currTypedDep.dep().index();
        if (govIndex != 0) {
            // Stanford CoreNLP produces a dependency relation between a verb and ROOT-0 which
            // is not token at all!
            Token govToken = tokens.get(govIndex - 1);
            Token depToken = tokens.get(depIndex - 1);

            sfAnnotator.createDependencyAnnotation(currTypedDep.reln(), govToken, depToken);
        }
    }
}

From source file:dependencies.TypedDependencyWrapper.java

License:Open Source License

/**
 * Constructor for class TypedDependencyWrapper
 */// w w w  .j  ava2  s.c  o m
public TypedDependencyWrapper(TypedDependency d) {
    govIndex = d.gov().index();
    govLabel = d.gov().value().toLowerCase();
    depIndex = d.dep().index();
    depLabel = d.dep().value().toLowerCase();
    relation = d.reln();
}