java - 斯坦福 NLP 语法关系类型？

Question

Soo... 该程序完全按照应有的方式运行，但我不明白该程序输出的花哨缩写是什么。

如果我输入'I wish today was a school day'，程序会输出

I/PRP : wish/VBP : nsubj
wish/VBP : day/NN : csubj
today/NN : wish/VBP : nmod:tmod
was/VBD : day/NN : cop
a/DT : day/NN : det
school/NN : day/NN : compound
day/NN : ROOT : root

我只想知道“cop”、“det”、“nsubj”、“csubj”等是什么；这样我就可以理解这些词之间的关系了。

String modelPath = null;
String taggerPath = null;
MaxentTagger tagger = null;
DependencyParser parser = null;
Properties props;
StanfordCoreNLP pipeline;

public ConversationEngine() {
    modelPath = DependencyParser.DEFAULT_MODEL;
    taggerPath = "edu/stanford/nlp/models/pos-tagger/english-left3words/english-left3words-distsim.tagger";
    tagger = new MaxentTagger(taggerPath);
    parser = DependencyParser.loadFromModelFile(modelPath);
    props = new Properties();
    props.setProperty("annotators", "tokenize, ssplit, pos, lemma, ner, parse, dcoref");
    pipeline = new StanfordCoreNLP(props);
}

public void parseSentence(String text) {
    String subject = "";
    DocumentPreprocessor tokenizer = new DocumentPreprocessor(new StringReader(text));
    for (List<HasWord> sentence : tokenizer) {
        List<TaggedWord> tagged = tagger.tagSentence(sentence);
        GrammaticalStructure gs = parser.predict(tagged);
        for (TypedDependency dep : gs.typedDependenciesEnhancedPlusPlus()) {
            System.out.println(dep.dep() + " : " + dep.gov() + " : " + dep.reln());

        }
    }
}

Answer 1

我相信你想要 dep.reln().toPrettyString()(参见 GrammaticalRelation#toPrettyString)

当您在 println 中调用 dep.reln() 时，这隐含为 dep.reln().toString() 并且javadocs 声明这

Returns short name (abbreviation) for this GrammaticalRelation.

(参见 GrammaticalRelation#toString)