static Client()
{
// Create a StanfordCoreNLPClient object with POS tagging, lemmatization, NER, parsing, and coreference resolution
java.util.Properties props = new java.util.Properties();
props.setProperty("annotators", "tokenize, ssplit, pos, lemma, ner, parse, dcoref");
props.setProperty("coref.algorithm", "neural");
ServerPipeline = new StanfordCoreNLPClient(props, Properties.Settings.Default.CoreNLP_ServerHost, Properties.Settings.Default.CoreNLP_ServerPort, ClientThreads);
}
static EnglishPOSExtractor()
{
var props = new java.util.Properties();
props.setProperty("annotators", "tokenize, ssplit, pos");
props.setProperty("ner.useSUTime", "0");
var curDir = Environment.CurrentDirectory;
Directory.SetCurrentDirectory($"{curDir}/english");
pipeline = new StanfordCoreNLP(props);
Directory.SetCurrentDirectory(curDir);
}
/// <summary>
/// Initialises CoreNLP pipeline to tokenise the input into sentences
/// </summary>
private void InitPipeline()
{
var jarRoot = @"C:\Users\Luca\Documents\University\General\UG3\SCC300\SCC300cs\SCC300cs\stanford-corenlp-3.7.0-models";
// Annotation pipeline configuration
var props = new java.util.Properties();
props.setProperty("annotators", "tokenize, ssplit");
props.setProperty("ner.useSUTime", "0");
var CurDir = Environment.CurrentDirectory;
Directory.SetCurrentDirectory(jarRoot); //change working dir to locate models
Directory.SetCurrentDirectory(CurDir);
}
// Sample from https://stanfordnlp.github.io/CoreNLP/corenlp-server.html
static void Main()
{
// creates a StanfordCoreNLP object with POS tagging, lemmatization, NER, parsing, and coreference resolution
var props = new java.util.Properties();
props.setProperty("annotators", "tokenize, ssplit, pos, lemma, ner, parse, dcoref");
StanfordCoreNLPClient pipeline = new StanfordCoreNLPClient(props, "http://localhost", 9000, 2);
// read some text in the text variable
var text = "Kosgi Santosh sent an email to Stanford University.";
// create an empty Annotation just with the given text
Annotation document = new Annotation(text);
// run all Annotators on this text
pipeline.annotate(document);
var sentences = document.get(sentencesAnnotationClass) as java.util.AbstractList;
foreach (CoreMap sentence in sentences)
{
var tokens = sentence.get(tokensAnnotationClass) as java.util.AbstractList;
Console.WriteLine("----");
foreach (CoreLabel token in tokens)
{
var word = token.get(textAnnotationClass);
var pos = token.get(partOfSpeechAnnotationClass);
var ner = token.get(namedEntityTagAnnotationClass);
Console.WriteLine("{0}\t[pos={1};\tner={2};", word, pos, ner);
}
}
}
/// <summary>
/// Annotation with SUTime
/// </summary>
public ScenarioAnnotator(IEmbeddingNetwork net)
{
//词向量模型
_net = net;
//annotate properites
_props = new java.util.Properties();
//refrenece https://stanfordnlp.github.io/CoreNLP/annotators.html
_props.setProperty("annotators",
//tokenize https://stanfordnlp.github.io/CoreNLP/tokenize.html
"tokenize, " +
//https://stanfordnlp.github.io/CoreNLP/cleanxml.html
//"cleanxml, " +
//ssplit https://stanfordnlp.github.io/CoreNLP/ssplit.html
"ssplit, " +
//part of speech https://stanfordnlp.github.io/CoreNLP/pos.html
"pos, " +
//lemma https://stanfordnlp.github.io/CoreNLP/lemma.html
"lemma, " +
//named entity recongnition https://stanfordnlp.github.io/CoreNLP/ner.html
"ner, " +
//depparse https://stanfordnlp.github.io/CoreNLP/parse.html
"depparse, " +
//Open Information Extraction https://stanfordnlp.github.io/CoreNLP/openie.html
"openie");
}
public void StanfordCore(string jarRoot = @"..\..\models")
{
var props = new java.util.Properties();
props.setProperty("annotators", "tokenize, ssplit, pos, lemma");
// props.setProperty("ner.useSUTime", "0");
var curDir = Environment.CurrentDirectory;
Directory.SetCurrentDirectory(jarRoot);
pipeline = new StanfordCoreNLP(props);
Directory.SetCurrentDirectory(curDir);
}
public StanfordCoreNLP()
{
var props = new java.util.Properties();
props.setProperty("annotators", "tokenize, ssplit, parse, sentiment");
var dictBackup = Environment.CurrentDirectory;
System.IO.Directory.SetCurrentDirectory(_modelsFolder);
_nlp = new edu.stanford.nlp.pipeline.StanfordCoreNLP(props);
System.IO.Directory.SetCurrentDirectory(dictBackup);
}
/// <summary inherit="yes"/>
public override JProperties GetOutputProperties()
{
JProperties properties = (defaultOutputProperties == null ? new JProperties() : new JProperties(defaultOutputProperties));
java.util.Enumeration propsEnum = props.keys();
while (propsEnum.hasMoreElements())
{
object obj = propsEnum.nextElement();
String value = (String)(props.get((String)obj));
properties.setProperty((String)obj, value);
}
return(properties);
}
public void Pos()
{
var sent = new Sentence("Lucy is in the sky with diamonds.");
var props = new java.util.Properties();
props.setProperty("ner.useSUTime", "0");
var nerTags = sent.nerTags(props);
Assert.AreEqual("PERSON", nerTags.get(0));
var firstPOSTag = sent.posTag(0);
Assert.AreEqual("NNP", firstPOSTag);
}
/// <summary>
///
/// </summary>
/// <returns></returns>
public bool Initialize()
{
if (this.Initialized)
{
return(true);
}
var props = new java.util.Properties();
// we tokenize and sentece split
props.setProperty("annotators", "tokenize, ssplit");
// don't separate words only when whitespace is encountered
// e.g. ***THIS IS... = *** + THIS + IS...
props.setProperty("tokenize.options", "whitespace=false");
// totally ignores parentheses and brackets - we don't care for them
// NOTE (twolf): this doesn't really work - so we take care of it in
// |QualifiedWords|
props.setProperty("tokenize.options", "normalizeOtherBrackets=false");
props.setProperty("tokenize.options", "normalizeParentheses=false");
// version 3.7.0 of CoreNLP supports splitting hypentated words
// yet has a whitelist of hyphenated words - so we let it do the work
// for us and refrain from really long words like:
// pg345.txt: two-pages-to-the-week-with-Sunday-squeezed-in-a-corner
props.setProperty("tokenize.options", "splitHyphenated=true");
// two or more newlines should be treated as a sentece break
// this is especially important for tables of contents
props.setProperty(
StanfordCoreNLP.NEWLINE_IS_SENTENCE_BREAK_PROPERTY, "two");
// ignore
props.setProperty("ssplit.tokenPatternsToDiscard", "\\p{Punct}");
try {
_pipeLine = new StanfordCoreNLP(props);
} catch (Exception) {
_pipeLine = null;
}
return(this.Initialized);
}
/// <summary>
/// Output an XML representation of the compiled code of the stylesheet, for purposes of
/// diagnostics and instrumentation
/// </summary>
/// <param name="destination">The destination for the diagnostic output</param>
public void Explain(XmlDestination destination) {
JConfiguration config = pss.getConfiguration();
JResult result = destination.GetResult(config.makePipelineConfiguration());
JProperties properties = new JProperties();
properties.setProperty("indent", "yes");
properties.setProperty("{http://saxon.sf.net/}indent-spaces", "2");
JReceiver receiver = config.getSerializerFactory().getReceiver(
result, config.makePipelineConfiguration(), properties);
JExpressionPresenter presenter = new JExpressionPresenter(config, receiver);
pss.explain(presenter);
}
/// <summary>Set a serialization property</summary>
/// <remarks>In the case of XSLT, properties set within the serializer override
/// any properties set in <c>xsl:output</c> declarations in the stylesheet.
/// Similarly, with XQuery, they override any properties set in the Query
/// prolog using <c>declare option saxon:output</c>.</remarks>
/// <example>
/// <code>
/// Serializer qout = new Serializer();
/// qout.SetOutputProperty(Serializer.METHOD, "xml");
/// qout.SetOutputProperty(Serializer.INDENT, "yes");
/// qout.SetOutputProperty(Serializer.SAXON_INDENT_SPACES, "1");
/// </code>
/// </example>
/// <param name="name">The name of the serialization property to be set</param>
/// <param name="value">The value to be set for the serialization property. May be null
/// to unset the property (that is, to set it back to the default value).</param>
public void SetOutputProperty(QName name, String value)
{
props.setProperty(name.ClarkName, value);
}
private java.util.Properties InitProperties()
{
var properties = new java.util.Properties();
properties.setProperty("parse.model", localModelPath + "edu/stanford/nlp/models/srparser/englishSR.ser.gz");
properties.setProperty("sentiment.model", localModelPath + "edu/stanford/nlp/models/sentiment/sentiment.ser.gz");
properties.setProperty("pos.model", localModelPath + "edu/stanford/nlp/models/pos-tagger/english-left3words/english-left3words-distsim.tagger");
properties.setProperty("ner.model", localModelPath + "edu/stanford/nlp/models/ner/english.all.3class.distsim.crf.ser.gz");
properties.setProperty("dcoref.demonym", localModelPath + "edu/stanford/nlp/models/dcoref/demonyms.txt");
properties.setProperty("dcoref.states", localModelPath + "edu/stanford/nlp/models/dcoref/state-abbreviations.txt");
properties.setProperty("dcoref.animate", localModelPath + "edu/stanford/nlp/models/dcoref/animate.unigrams.txt");
properties.setProperty("dcoref.inanimate", localModelPath + "edu/stanford/nlp/models/dcoref/inanimate.unigrams.txt");
properties.setProperty("dcoref.big.gender.number", localModelPath + "edu/stanford/nlp/models/dcoref/gender.data.gz");
properties.setProperty("dcoref.countries", localModelPath + "edu/stanford/nlp/models/dcoref/countries");
properties.setProperty("dcoref.states.provinces", localModelPath + "edu/stanford/nlp/models/dcoref/statesandprovinces");
properties.setProperty("dcoref.singleton.model", localModelPath + "edu/stanford/nlp/models/dcoref/singleton.predictor.ser");
properties.setProperty("annotators", "tokenize, ssplit, pos, parse, lemma, ner, sentiment");
properties.setProperty("tokenize.language", "en");
properties.setProperty("ner.useSUTime", "0");
properties.setProperty("sutime.binders", "0");
properties.setProperty("sutime.rules", localModelPath + "edu/stanford/nlp/models/sutime/defs.sutime.txt, " + localModelPath + "edu/stanford/nlp/models/sutime/english.sutime.txt");
return(properties);
}
static ChinesePOSExtractor()
{
var props = new java.util.Properties();
props.setProperty("annotators", "segment, ssplit, pos");
props.setProperty("customAnnotatorClass.segment", "edu.stanford.nlp.pipeline.ChineseSegmenterAnnotator");
props.setProperty("segment.model", "edu/stanford/nlp/models/segmenter/chinese/ctb.gz");
props.setProperty("segment.sighanCorporaDict", "edu/stanford/nlp/models/segmenter/chinese");
props.setProperty("segment.serDictionary", "edu/stanford/nlp/models/segmenter/chinese/dict-chris6.ser.gz");
props.setProperty("segment.sighanPostProcessing", "true");
//sentence split
props.setProperty("ssplit.boundaryTokenRegex", "[.]|[!?]+|[。]|[！？]+");
//pos
props.setProperty("pos.model", "edu/stanford/nlp/models/pos-tagger/chinese-distsim/chinese-distsim.tagger");
//ner
props.setProperty("ner.model", "edu/stanford/nlp/models/ner/chinese.misc.distsim.crf.ser.gz");
props.setProperty("ner.applyNumericClassifiers", "false");
props.setProperty("ner.useSUTime", "false");
//# parse
props.setProperty("parse.model", "edu/stanford/nlp/models/lexparser/chineseFactored.ser.gz");
pipeline = new StanfordCoreNLP(props);
}
/// <summary>Set a serialization property</summary>
/// <remarks>In the case of XSLT, properties set within the serializer override
/// any properties set in <c>xsl:output</c> declarations in the stylesheet.
/// Similarly, with XQuery, they override any properties set in the Query
/// prolog using <c>declare option saxon:output</c>.</remarks>
/// <example>
/// <code>
/// Serializer qout = new Serializer();
/// qout.SetOutputProperty(Serializer.METHOD, "xml");
/// qout.SetOutputProperty(Serializer.INDENT, "yes");
/// qout.SetOutputProperty(Serializer.SAXON_INDENT_SPACES, "1");
/// </code>
/// </example>
/// <param name="name">The name of the serialization property to be set</param>
/// <param name="value">The value to be set for the serialization property. May be null
/// to unset the property (that is, to set it back to the default value).</param>
public void SetOutputProperty(QName name, String value)
{
props.setProperty(name.ClarkName, value);
serializer.setOutputProperty(net.sf.saxon.s9api.Serializer.getProperty(name.UnderlyingQName()), value);
}
public void SetProperty(string key, string value)
{
underlyingModel.setProperty(key, value);
}
//使用nlp將文章分析後回傳key
private List<string> nlp(string sentence)
{
List<string> return_key = new List<string>();
string Relay_file = ".\\xml";
string Relay_name = "Relay" + ".xml";
string Relay_path = Relay_file + "\\" + Relay_name;
// Path to the folder with models extracted from `stanford-corenlp-3.4-models.jar`
var jarRoot = @"stanford-corenlp-3.5.2-models\\";
// Annotation pipeline configuration
var props = new java.util.Properties();
props.setProperty("ner.useSUTime", "false");
props.setProperty("annotators", "tokenize, ssplit, pos, lemma, ner, parse, dcoref");
props.setProperty("sutime.binders", "0");
// We should change current directory, so StanfordCoreNLP could find all the model files automatically
var curDir = Environment.CurrentDirectory;
System.IO.Directory.SetCurrentDirectory(jarRoot);
var pipeline = new StanfordCoreNLP(props);
System.IO.Directory.SetCurrentDirectory(curDir);
// Annotation
var annotation = new Annotation(sentence);
pipeline.annotate(annotation);
//輸出nlp分析結果至Relay.xml
FileOutputStream os = new FileOutputStream(new File(Relay_file, Relay_name));
pipeline.xmlPrint(annotation, os);
os.close();
//呼叫ner將單字組合為有意義的key組裝
foreach (string k in ner(Relay_path))
{
return_key.Add(k);
}
return return_key;
}