Example usage for edu.stanford.nlp.tagger.maxent MaxentTagger tagSentence

Introduction

In this page you can find the example usage for edu.stanford.nlp.tagger.maxent MaxentTagger tagSentence.

Prototype

public List<TaggedWord> tagSentence(List<? extends HasWord> sentence, boolean reuseTags) 

Source Link

Document

Returns a new Sentence that is a copy of the given sentence with all the words tagged with their part-of-speech.

Usage

From source file:ie.pars.bnc.preprocess.ProcessNLP.java

License:Open Source License

/**
 *
 * @param inputStreamFile/*  www .  jav a2 s.com*/
 * @param morphology
 * @param posTagger
 * @param parser
 * @return
 * @throws Exception
 */
public static StringBuilder parseBNCXML(InputStream inputStreamFile, Morphology morphology,
        MaxentTagger posTagger, ParserGrammar parser) throws Exception {
    StringBuilder results = new StringBuilder();
    int counterSent = 0;
    List<List<List<WordLemmaTag>>> parseBNCXMLTokenized = parseBNCXMLTokenized(inputStreamFile);
    for (List<List<WordLemmaTag>> xparseBNCXMLL : parseBNCXMLTokenized) {
        results.append("<p>\n");
        for (List<WordLemmaTag> para : xparseBNCXMLL) {
            if (counterSent++ % 20 == 0) {
                System.out.print(".");
            }
            results.append("<s>\n");
            List<TaggedWord> tagSentence = posTagger.tagSentence(para, true);

            Tree parseTree = parser.parse(tagSentence);

            GrammaticalStructure gs = parser.getTLPParams().getGrammaticalStructure(parseTree,
                    parser.treebankLanguagePack().punctuationWordRejectFilter(),
                    parser.getTLPParams().typedDependencyHeadFinder());

            Collection<TypedDependency> deps = gs.typedDependenciesCollapsedTree();
            SemanticGraph depTree = new SemanticGraph(deps);

            for (int i = 0; i < tagSentence.size(); ++i) {

                int head = -1;
                String deprel = null;
                //                    if (depTree != null) {
                Set<Integer> rootSet = depTree.getRoots().stream().map(IndexedWord::index)
                        .collect(Collectors.toSet());
                IndexedWord node = depTree.getNodeByIndexSafe(i + 1);
                if (node != null) {
                    List<SemanticGraphEdge> edgeList = depTree.getIncomingEdgesSorted(node);
                    if (!edgeList.isEmpty()) {
                        assert edgeList.size() == 1;
                        head = edgeList.get(0).getGovernor().index();
                        deprel = edgeList.get(0).getRelation().toString();
                    } else if (rootSet.contains(i + 1)) {
                        head = 0;
                        deprel = "ROOT";
                    }
                }
                //     }

                // Write the token
                TaggedWord lexHead = null;
                if (head > 0) {
                    lexHead = tagSentence.get(head - 1);
                }
                results.append(line(i + 1, tagSentence.get(i), morphology, head, deprel, lexHead)).append("\n");
            }
            results.append("</s>\n");
        }
        results.append("</p>\n");
    }
    System.out.println("");
    inputStreamFile.close();

    return results;
}

---