Stanford CoreNLP

下载Stanford CoreNLP的压缩包，地址：https://stanfordnlp.github.io/CoreNLP/api.html
新建java项目，引入压缩包里的jar包（project右键——>build Path——>Configure BuildPath——>libraries——>add external jars）选中压缩包文件夹里的jar文件，引入即可。

新建class文件：

package com.ww.corenlp;
import java.util.List;
import java.util.Map;
import java.util.Properties;

import edu.stanford.nlp.dcoref.CorefChain;
import edu.stanford.nlp.dcoref.CorefCoreAnnotations.CorefChainAnnotation;
import edu.stanford.nlp.ling.CoreAnnotations.LemmaAnnotation;
import edu.stanford.nlp.ling.CoreAnnotations.NamedEntityTagAnnotation;
import edu.stanford.nlp.ling.CoreAnnotations.PartOfSpeechAnnotation;
import edu.stanford.nlp.ling.CoreAnnotations.SentencesAnnotation;
import edu.stanford.nlp.ling.CoreAnnotations.TextAnnotation;
import edu.stanford.nlp.ling.CoreAnnotations.TokensAnnotation;
import edu.stanford.nlp.ling.CoreLabel;
import edu.stanford.nlp.pipeline.Annotation;
import edu.stanford.nlp.pipeline.StanfordCoreNLP;
import edu.stanford.nlp.semgraph.SemanticGraph;
import edu.stanford.nlp.semgraph.SemanticGraphCoreAnnotations.CollapsedCCProcessedDependenciesAnnotation;
import edu.stanford.nlp.sentiment.SentimentCoreAnnotations;
import edu.stanford.nlp.trees.Tree;
import edu.stanford.nlp.trees.TreeCoreAnnotations.TreeAnnotation;
import edu.stanford.nlp.util.CoreMap;

public class TestNLP {
    public static void main(String[] args) {
        // creates a StanfordCoreNLP object, with POS tagging, lemmatization, NER, parsing, and coreference resolution
        Properties props = new Properties();
        props.put("annotators", "tokenize, ssplit, pos, lemma, ner, parse, dcoref");
        StanfordCoreNLP pipeline = new StanfordCoreNLP(props);
        // read some text in the text variable
        String text = "Mrs. Clinton previously worked for Mr. Obama, but she is  now distancing herself from him";
        
        // create an empty Annotation just with the given text
        Annotation document = new Annotation(text);
        
        // run all Annotators on this text
        pipeline.annotate(document);
        
        // these are all the sentences in this document
        // a CoreMap is essentially a Map that uses class objects as keys and has values with custom types
        List sentences = document.get(SentencesAnnotation.class);
        
        System.out.println("word\t  pos\t  lemma\t  ner");
        for(CoreMap sentence: sentences) {
             // traversing the words in the current sentence
             // a CoreLabel is a CoreMap with additional token-specific methods
            for (CoreLabel token: sentence.get(TokensAnnotation.class)) {
                // this is the text of the token
                String word = token.get(TextAnnotation.class);
                // this is the POS tag of the token
                String pos = token.get(PartOfSpeechAnnotation.class);
                // this is the NER label of the token
                String ne = token.get(NamedEntityTagAnnotation.class);
                String lemma = token.get(LemmaAnnotation.class);
               
                System.out.println(word+"\t"+pos+"\t"+lemma+"\t"+ne);
            }
            // this is the parse tree of the current sentence
            // 句子的解析树  
            Tree tree = sentence.get(TreeAnnotation.class);
            System.out.println("\nparse tree:");
            tree.pennPrint();  
            // this is the Stanford dependency graph of the current sentence
            // 句子的依赖图  
            SemanticGraph dependencies = sentence.get(CollapsedCCProcessedDependenciesAnnotation.class);
            System.out.println("\ndependencies:");
            System.out.println(dependencies.toString(SemanticGraph.OutputFormat.LIST));  
        }
        // This is the coreference link graph
        // Each chain stores a set of mentions that link to each other,
        // along with a method for getting the most representative mention
        // Both sentence and token offsets start at 1!
        Map graph = document.get(CorefChainAnnotation.class);
    }
}

针对Advances in natural language processing论文中提到的句子做处理：

Mrs. Clinton previously worked for Mr. Obama, but she is  now distancing herself from him

可以在官方的在线demo上尝试：

http://nlp.stanford.edu:8080/parser/index.jsp

得到结果为：

word      pos     lemma   ner
Mrs.    NNP Mrs.    O
Clinton NNP Clinton PERSON
previously  RB  previously  DATE
worked  VBD work    O
for IN  for O
Mr. NNP Mr. O
Obama   NNP Obama   PERSON
,   ,   ,   O
but CC  but O
she PRP she O
is  VBZ be  O
now RB  now DATE
distancing  VBG distance    O
herself PRP herself O
from    IN  from    O
him PRP he  O

parse tree:
(ROOT
  (FRAG
    (S
      (S
        (NP (NNP Mrs.) (NNP Clinton))
        (ADVP (RB previously))
        (VP (VBD worked)
          (PP (IN for)
            (NP (NNP Mr.) (NNP Obama)))))
      (, ,)
      (CC but)
      (S
        (NP (PRP she))
        (VP (VBZ is)
          (ADVP (RB now))
          (VP (VBG distancing)
            (NP (PRP herself))
            (PP (IN from)
              (NP (PRP him)))))))))

dependencies:
root(ROOT-0, worked-4)
compound(Clinton-2, Mrs.-1)
nsubj(worked-4, Clinton-2)
advmod(worked-4, previously-3)
case(Obama-7, for-5)
compound(Obama-7, Mr.-6)
nmod:for(worked-4, Obama-7)
punct(worked-4, ,-8)
cc(worked-4, but-9)
nsubj(distancing-13, she-10)
aux(distancing-13, is-11)
advmod(distancing-13, now-12)
conj:but(worked-4, distancing-13)
dobj(distancing-13, herself-14)
case(him-16, from-15)
nmod:from(distancing-13, him-16)

同论文中效果对比：

Paste_Image.png

引用：REVIEW
Advances in natural language processing Julia Hirschberg 1 and Christopher D. Manning 2,3
以上：
祝好！