斯坦福 CoreNLP

Stanford CoreNLP 是一种流行的自然语言处理工具包，支持许多核心 NLP 任务。

要下载并安装该程序，请下载发行包并在类路径中包含必要的*.jar 文件，或者从 Maven 中心添加依赖项。有关详细信息，请参阅下载页面 。例如：

curl http://nlp.stanford.edu/software/stanford-corenlp-full-2015-12-09.zip -o corenlp.zip
unzip corenlp.zip
cd corenlp
export CLASSPATH="$CLASSPATH:`pwd`/*

运行 CoreNLP 工具有三种支持的方法：（1）使用基本完全可自定义的 API ，（2）使用 Simple CoreNLP API，或（3）使用 CoreNLP 服务器 。下面给出每个的简单使用示例。作为一个激励用例，这些例子将用于预测句子的句法分析。

1. CoreNLP API

   public class CoreNLPDemo {
     public static void main(String[] args) {
   
       // 1. Set up a CoreNLP pipeline. This should be done once per type of annotation,
       //    as it's fairly slow to initialize.
       // creates a StanfordCoreNLP object, with POS tagging, lemmatization, NER, parsing, and coreference resolution 
       Properties props = new Properties();
       props.setProperty("annotators", "tokenize, ssplit, parse");
       StanfordCoreNLP pipeline = new StanfordCoreNLP(props);
   
       // 2. Run the pipeline on some text.
       // read some text in the text variable
       String text = "the quick brown fox jumped over the lazy dog"; // Add your text here!
       // create an empty Annotation just with the given text
       Annotation document = new Annotation(text);
       // run all Annotators on this text
       pipeline.annotate(document);
   
       // 3. Read off the result
       // Get the list of sentences in the document
       List<CoreMap> sentences = document.get(CoreAnnotations.SentencesAnnotation.class);
       for (CoreMap sentence : sentences) {
         // Get the parse tree for each sentence
         Tree parseTree = sentence.get(TreeAnnotations.TreeAnnotation.class);
         // Do something interesting with the parse tree!
         System.out.println(parseTree);
       }
   
     }
   }
   

2. 简单的 CoreNLP

   public class CoreNLPDemo {
     public static void main(String[] args) {
       String text = "The quick brown fox jumped over the lazy dog");  // your text here!
       Document document = new Document(text);  // implicitly runs tokenizer
       for (Sentence sentence : document.sentences()) {
         Tree parseTree = sentence.parse();  // implicitly runs parser
         // Do something with your parse tree!
         System.out.println(parseTree);
       }
     } 
   }
   

3. CoreNLP 服务器

   使用以下命令启动服务器（适当地设置类路径）：

   java -mx4g -cp "*" edu.stanford.nlp.pipeline.StanfordCoreNLPServer [port] [timeout]
   

   获取给定注释器集的 JSON 格式输出，并将其打印到标准输出：

    wget --post-data 'The quick brown fox jumped over the lazy dog.' 'localhost:9000/?properties={"annotators":"tokenize,ssplit,parse","outputFormat":"json"}' -O -
   

   要从 JSON 获取我们的解析树，我们可以将 JSON 导航到 sentences[i].parse。