public void ThreeInterestingDates()
{
var pipeline = new AnnotationPipeline();
pipeline.addAnnotator(new TokenizerAnnotator(false));
pipeline.addAnnotator(new WordsToSentencesAnnotator(false));
var filePath = Files.CoreNlp.Models("pos-tagger/english-left3words-distsim.tagger");
var tagger = new MaxentTagger(filePath);
pipeline.addAnnotator(new POSTaggerAnnotator(tagger));
var props = Java.Props(new Dictionary <string, string>
{
{ "sutime.binders", "0" },
{
"sutime.rules", string.Join(",",
Files.CoreNlp.Models("sutime/defs.sutime.txt"),
Files.CoreNlp.Models("sutime/english.sutime.txt"),
Files.CoreNlp.Models("sutime/english.holidays.sutime.txt"))
}
});
pipeline.addAnnotator(new TimeAnnotator("sutime", props));
var text = "Three interesting dates are 18 Feb 1997, the 20th of july and 4 days from today.";
var annotation = new Annotation(text);
annotation.set(Java.GetAnnotationClass <CoreAnnotations.DocDateAnnotation>(), "2013-07-14");
pipeline.annotate(annotation);
TestContext.Out.WriteLine(annotation.get(Java.GetAnnotationClass <CoreAnnotations.TextAnnotation>()));
var timexAnnsAll = (java.util.ArrayList)annotation.get(Java.GetAnnotationClass <TimeAnnotations.TimexAnnotations>());
Assert.Greater(timexAnnsAll.size(), 0);
foreach (CoreMap cm in timexAnnsAll)
{
var tokens = (java.util.List)cm.get(Java.GetAnnotationClass <CoreAnnotations.TokensAnnotation>());
Assert.Greater(tokens.size(), 0);
var first = tokens.get(0);
var last = tokens.get(tokens.size() - 1);
var time = (TimeExpression)cm.get(Java.GetAnnotationClass <TimeExpression.Annotation>());
Assert.IsNotNull(time, "Time expression is null");
TestContext.Out.WriteLine($"{cm} [from char offset '{first}' to '{last}'] --> {time.getTemporal()}");
}
}
public static void CoreNlpClient()
{
// creates a StanfordCoreNLP object with POS tagging, lemmatization, NER, parsing, and coreference resolution
var props = new Properties();
props.setProperty("annotators", "tokenize, ssplit, pos, lemma, ner, parse, dcoref");
StanfordCoreNLPClient pipeline = new(props, "http://localhost", 9000, 2);
// read some text in the text variable
var text = "Kosgi Santosh sent an email to Stanford University.";
// create an empty Annotation just with the given text
Annotation document = new(text);
// run all Annotators on this text
pipeline.annotate(document);
var sentencesAnnotationClass = Java.GetAnnotationClass <CoreAnnotations.SentencesAnnotation>();
var tokensAnnotationClass = Java.GetAnnotationClass <CoreAnnotations.TokensAnnotation>();
var textAnnotationClass = Java.GetAnnotationClass <CoreAnnotations.TextAnnotation>();
var partOfSpeechAnnotationClass = Java.GetAnnotationClass <CoreAnnotations.PartOfSpeechAnnotation>();
var namedEntityTagAnnotationClass = Java.GetAnnotationClass <CoreAnnotations.NamedEntityTagAnnotation>();
var sentences = sentencesAnnotationClass.getClasses().Select(document.get).ToList();
foreach (CoreMap sentence in sentences)
{
var tokens = (AbstractList)sentence.get(tokensAnnotationClass);
Console.WriteLine("----");
foreach (CoreLabel token in tokens)
{
var word = token.get(textAnnotationClass);
var pos = token.get(partOfSpeechAnnotationClass);
var ner = token.get(namedEntityTagAnnotationClass);
Console.WriteLine($"{word}\t[pos={pos};\tner={ner};]");
}
}
}