简单的 CoreNLP - 如何将所有名词放入一个数组？答案

【问题标题】：Simple CoreNLP - how to get all the nouns to an array?简单的 CoreNLP - 如何将所有名词放入一个数组？
【发布时间】：2017-06-10 05:14:29
【问题描述】：

我正在使用斯坦福简单 NLP。我需要将所有名词值获取到 nounPhrases 数组。 me() 方法给我的输出如下：

The parse of the sentence 'I like java and python' is (ROOT (S (NP (PRP I)) (VP (VBP like) (NP (NN java) (CC and) (NN python)))))

这是我的方法

public String s = "I like java and python";

public static Set<String> nounPhrases = new HashSet<>();

public void me() {

    Document doc = new Document(" " + s);
    for (Sentence sent : doc.sentences()) {

        System.out.println("The parse of the sentence '" + sent + "' is " + sent.parse());

        if (sent.parse().equals("NN") || sent.parse().equals("NNS") || sent.parse().equals("NNP")
                || sent.parse().equals("NNPS")) {

            // I need to assign all nouns to the array nounPhrases

        }

    }
}

我不确定我的 if 条件是对还是错，因为我是斯坦福 NLP 的新手。请帮我把我的名词放到这个数组中。

我得到了 URL 下面的示例代码表单，并对其进行了一些自定义。

Simple CoreNLP

【问题讨论】：

标签： java arrays parsing stanford-nlp

【解决方案1】：

如果有人需要此解决方案的完整和最新版本，这里是：

import java.util.HashSet;
import java.util.Properties;
import java.util.Set;

import edu.stanford.nlp.pipeline.CoreDocument;
import edu.stanford.nlp.pipeline.CoreSentence;
import edu.stanford.nlp.pipeline.StanfordCoreNLP;


public class BasicPipelineExample4 {

  public static String text = "Joe Smith was born in California. "+
  "Study studying studied. " +
  "In 2017, he went to Paris, France in the summer. " +
  "His flight left at 3:00pm on July 10th, 2017. " +
  "After eating some escargot for the first time, Joe said, \"That was delicious!\" " +
  "He sent a postcard to his sister Jane Smith. " +
  "He is ok. " +
  "Simple, right? Remove removed removing was were is are element at given gave give index, insert it at desired index. Let's see if it works for the second test case."+
  "He is ok to go now. " +
  "After hearing about Joe's trip, Jane decided she might go to France one day.";

public static void main(String[] args) {
    Properties props = new Properties();
    // set the list of annotators to run
    props.setProperty("annotators", "tokenize,ssplit,pos,parse");
    // set a property for an annotator, in this case the coref annotator is being
    // set to use the neural algorithm
    props.setProperty("coref.algorithm", "neural");
    // build pipeline
    StanfordCoreNLP pipeline = new StanfordCoreNLP(props);
    // create a document object
    CoreDocument doc = new CoreDocument(text);
    // annnotate the document
    pipeline.annotate(doc);

    Set<String> nounPhrases = new HashSet<>();

    for (CoreSentence sent : doc.sentences()) {

        System.out.println("The parse of the sentence '" + sent + "' is " + sent.constituencyParse());
        // Iterate over every word in the sentence
        for (int i = 0; i < sent.tokens().size(); i++) {
            // Condition: if the word is a noun (posTag starts with "NN")
            if (sent.posTags() != null && sent.posTags().get(i) != null && sent.posTags().get(i).contains("NN")) {
                // Put the word into the Set
                nounPhrases.add(sent.tokens().get(i).originalText());
            }
        }
    }

    System.out.println("Nouns: " + nounPhrases);

}

}

【讨论】：

【解决方案2】：

您的情况几乎是正确的。您想要每个具有包含“NN”的 POS 标签的单词，即每个名词。要检查每个单词的 POS 标签，您必须遍历句子中的每个单词。根据您的代码，它可能如下所示：

public String s = "I like java and python";

public static Set<String> nounPhrases = new HashSet<>();

public void me() {

    Document doc = new Document(" " + s);
    for (Sentence sent : doc.sentences()) {

        System.out.println("The parse of the sentence '" + sent + "' is " + sent.parse());
        //Iterate over every word in the sentence
        for(int i = 0; i < sent.words().size(); i++) {
            //Condition: if the word is a noun (posTag starts with "NN")
            if (sent.posTag(i) != null && sent.posTag(i).contains("NN")) {
                //Put the word into the Set
                nounPhrases.add(sent.word(i));
            }
        }
    }
}

【讨论】：