static void Main()
{
// Path to the folder with models extracted from `stanford-corenlp-3.9.1-models.jar`
var jarRoot = @"..\..\..\..\data\paket-files\nlp.stanford.edu\stanford-corenlp-full-2018-10-05\models";
// Text for processing
var text = "Kosgi Santosh sent an email to Stanford University. He didn't get a reply.";
// Annotation pipeline configuration
var props = new Properties();
props.setProperty("annotators", "tokenize, ssplit, pos, lemma, parse, ner,dcoref");
props.setProperty("ner.useSUTime", "0");
// We should change current directory, so StanfordCoreNLP could find all the model files automatically
var curDir = Environment.CurrentDirectory;
Directory.SetCurrentDirectory(jarRoot);
var pipeline = new StanfordCoreNLP(props);
Directory.SetCurrentDirectory(curDir);
// Annotation
var annotation = new Annotation(text);
pipeline.annotate(annotation);
// Result - Pretty Print
using (var stream = new ByteArrayOutputStream())
{
pipeline.prettyPrint(annotation, new PrintWriter(stream));
Console.WriteLine(stream.toString());
stream.close();
}
}
//-------------------------------------------------------------------------------------- Stanford Core NLP -----------------------------------------
//-- Better for Entity recognition
public static void buildPipeline(string text)
{//https://interviewbubble.com/getting-started-with-stanford-corenlp-a-stanford-corenlp-tutorial/
// Path to the folder with models extracted from `stanford-corenlp-3.7.0-models.jar`
var jarRoot = @"..\..\..\..\data\paket-files\nlp.stanford.edu\stanford-corenlp-full-2016-10-31\models";
// creates a StanfordCoreNLP object, with POS tagging, lemmatization,
// NER, parsing, and coreference resolution
Properties props = new Properties();
props.setProperty("annotators", "tokenize, ssplit, pos, lemma, ner, parse, dcoref");
StanfordCoreNLP pipeline = new StanfordCoreNLP(props);
// create an empty Annotation just with the given text
Annotation document = new Annotation(text);
// run all Annotators on this text
pipeline.annotate(document);
//Finished processing the document here
// Result - Pretty Print
using (var stream = new ByteArrayOutputStream())
{
pipeline.prettyPrint(document, new PrintWriter(stream));
Debug.WriteLine(stream.toString());
stream.close();
}
}
public String testFreemarker()
{
Assembly _assembly;
_assembly = Assembly.GetExecutingAssembly();
//Console.WriteLine(_assembly.
try{
Configuration cfg = new Configuration();
cfg.setDirectoryForTemplateLoading(new File("template"));
//cfg.setDirectoryForTemplateLoading(new File(""));
cfg.setObjectWrapper(new DefaultObjectWrapper());
Template temp = cfg.getTemplate("c.ftl");
Map root = new HashMap();
root.put("codeGen", this);
ByteArrayOutputStream outputStream = new ByteArrayOutputStream();
Writer output = new OutputStreamWriter(outputStream);
temp.process(root, output);
output.flush();
//System.Console.WriteLine(outputStream.toString());
return(outputStream.toString());
}
catch (IOException exception) {
} catch (TemplateException exception) {
}
return("");
}
public void StanfordCoreNlpDemoThatChangeCurrentDirectory()
{
const string Text = "Kosgi Santosh sent an email to Stanford University. He didn't get a reply.";
// Annotation pipeline configuration
var props = new Properties();
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
var curDir = Environment.CurrentDirectory;
Directory.SetCurrentDirectory(Config.JarRoot);
var pipeline = new edu.stanford.nlp.pipeline.StanfordCoreNLP(props);
Directory.SetCurrentDirectory(curDir);
// Annotation
var annotation = new Annotation(Text);
pipeline.annotate(annotation);
// Result - Pretty Print
using (var stream = new ByteArrayOutputStream())
{
pipeline.prettyPrint(annotation, new PrintWriter(stream));
Console.WriteLine(stream.toString());
}
this.CustomAnnotationPrint(annotation);
}
public ParsedStatementFactory.ParseResult ParseStatement(string input)
{
var sent2Reader = new StringReader(input);
var rawWords2 = _tokenizerFactory.getTokenizer(sent2Reader).tokenize();
var parse = _lp.apply(rawWords2);
var gs = _structureFactory.newGrammaticalStructure(parse);
var tdl = gs.typedDependenciesCCprocessed();
//System.Console.WriteLine("newGrammaticalStructure:\n{0}\n", gs);
//System.Console.WriteLine("typedDependenciesCCprocessed:\n{0}\n", tdl);
//var tp = new TreePrint("penn,typedDependenciesCollapsed");
//tp.printTree(parse);
//return new ParsedStatement(parse);
var xmlTreePrint = new TreePrint("xmlTree, dependencies", "xml, collapsedDependencies", _tlp);
var stream = new ByteArrayOutputStream();
xmlTreePrint.printTree(parse, new PrintWriter(stream));
string xmlOutput = stream.toString() + "</s>";
//System.Console.WriteLine("xml:\n{0}\n", xmlOutput);
return(ParsedStatementFactory.CreateParsedStatement(xmlOutput));
//System.Console.WriteLine("TreePrint: \n{0}\n", parse);
}
public String Read(File file) {
ByteArrayOutputStream buffer = new ByteArrayOutputStream();
InputStream source = new FileInputStream(file);
byte[] chunk = new byte[1024];
int count = 0;
while((count = source.Read(chunk)) != -1) {
buffer.write(chunk, 0, count);
}
return buffer.toString("UTF-8");
}
public NERActor()
{
Receive <string>(r =>
{
var req = JsonConvert.DeserializeObject <SharedMessages.NERRequest>(r);
var resp = new SharedMessages.NERResponse();
foreach (var l in req.linesToProcess)
{
var res = string.Format("<Results>{0}</Results>\n", Program.classifier.classifyWithInlineXML(l)).Replace("&", "(amp)");
XDocument xdoc = XDocument.Parse(res);
resp.results.Add(xdoc.Root.Elements().Select(z => z.Value).ToList());
}
resp.section = req.section;
resp.id = req.id;
resp.feed = req.feed;
Sender.Tell("ner:" + JsonConvert.SerializeObject(resp));
});
Receive <int>(r =>
{
var annotation = new Annotation(r.ToString());
Program.pipeline.annotate(annotation);
var NERs = annotation.get(typeof(edu.stanford.nlp.ling.CoreAnnotations.NamedEntityTagAnnotation));
if (NERs != null)
{
foreach (Annotation ner in NERs as ArrayList)
{
Console.WriteLine("NER: " + ner);
}
}
// 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
var sentences = annotation.get(typeof(edu.stanford.nlp.ling.CoreAnnotations.SentencesAnnotation));
if (sentences != null)
{
foreach (Annotation sentence in sentences as ArrayList)
{
Console.WriteLine("Sent: " + sentence);
}
}
// Result - Pretty Print
using (var stream = new ByteArrayOutputStream())
{
var pw = new PrintWriter(stream);
Program.pipeline.prettyPrint(annotation, pw);
Console.WriteLine(stream.toString());
stream.close();
}
});
}
public void StanfordCoreNlpDemoManualConfiguration()
{
Console.WriteLine(Environment.CurrentDirectory);
const string Text = "Kosgi Santosh sent an email to Stanford University. He didn't get a reply.";
// Annotation pipeline configuration
var props = new Properties();
props.setProperty("annotators", "tokenize, ssplit, pos, lemma, ner, parse, dcoref");
props.setProperty("pos.model", Config.GetModel(@"pos-tagger\english-bidirectional\english-bidirectional-distsim.tagger"));
props.setProperty("ner.model", Config.GetModel(@"ner\english.all.3class.distsim.crf.ser.gz"));
props.setProperty("parse.model", Config.GetModel(@"lexparser\englishPCFG.ser.gz"));
props.setProperty("dcoref.demonym", Config.GetModel(@"dcoref\demonyms.txt"));
props.setProperty("dcoref.states", Config.GetModel(@"dcoref\state-abbreviations.txt"));
props.setProperty("dcoref.animate", Config.GetModel(@"dcoref\animate.unigrams.txt"));
props.setProperty("dcoref.inanimate", Config.GetModel(@"dcoref\inanimate.unigrams.txt"));
props.setProperty("dcoref.male", Config.GetModel(@"dcoref\male.unigrams.txt"));
props.setProperty("dcoref.neutral", Config.GetModel(@"dcoref\neutral.unigrams.txt"));
props.setProperty("dcoref.female", Config.GetModel(@"dcoref\female.unigrams.txt"));
props.setProperty("dcoref.plural", Config.GetModel(@"dcoref\plural.unigrams.txt"));
props.setProperty("dcoref.singular", Config.GetModel(@"dcoref\singular.unigrams.txt"));
props.setProperty("dcoref.countries", Config.GetModel(@"dcoref\countries"));
props.setProperty("dcoref.extra.gender", Config.GetModel(@"dcoref\namegender.combine.txt"));
props.setProperty("dcoref.states.provinces", Config.GetModel(@"dcoref\statesandprovinces"));
props.setProperty("dcoref.singleton.predictor", Config.GetModel(@"dcoref\singleton.predictor.ser"));
props.setProperty("dcoref.big.gender.number", Config.GetModel(@"dcoref\gender.data.gz"));
var sutimeRules = new[] {
Config.GetModel(@"sutime\defs.sutime.txt"),
Config.GetModel(@"sutime\english.holidays.sutime.txt"),
Config.GetModel(@"sutime\english.sutime.txt")
};
props.setProperty("sutime.rules", String.Join(",", sutimeRules));
props.setProperty("sutime.binders", "0");
var pipeline = new edu.stanford.nlp.pipeline.StanfordCoreNLP(props);
// Annotation
var annotation = new Annotation(Text);
pipeline.annotate(annotation);
// Result - Pretty Print
using (var stream = new ByteArrayOutputStream())
{
pipeline.prettyPrint(annotation, new PrintWriter(stream));
Console.WriteLine(stream.toString());
}
this.CustomAnnotationPrint(annotation);
}
public NlpResult DeserializeInput(StanfordCoreNLP pipeline, NlpResult nlpResult, string stringInput)
{
// Annotation
var annotation = new Annotation(stringInput);
pipeline.annotate(annotation);
// Result - Pretty Print
using (var stream = new ByteArrayOutputStream())
{
pipeline.jsonPrint(annotation, new PrintWriter(stream));
_jsonContentProvider.PopulateFromString(nlpResult, stream.toString());
Debug.WriteLine(stream.toString());
stream.close();
}
return(nlpResult);
}
public static String Parse_Data(String sent, StanfordCoreNLP pipeline1)
{// extract meaningful words from user query
// Text for processing
var text = sent;
// Annotation
var annotation = new edu.stanford.nlp.pipeline.Annotation(text);
pipeline1.annotate(annotation);
// Result - Pretty Print
string output;
using (var stream = new ByteArrayOutputStream())
{
pipeline1.prettyPrint(annotation, new PrintWriter(stream));
System.Console.WriteLine(" it's stanford time ");
output = stream.toString();
stream.close();
}
return(output);
}
public static AnnotationObject Annotate(string content)
{
// Annotation
var annotation = new Annotation(content);
Pipeline.annotate(annotation);
// Result - Print
using var stream = new ByteArrayOutputStream();
Pipeline.jsonPrint(annotation, new PrintWriter(stream));
//-----
string serialized = stream.toString().Replace("\n", "");
var deserialized = Newtonsoft.Json.JsonConvert.DeserializeObject <AnnotationObject>(serialized);
//-----
stream.close();
return(deserialized);
}
public void CustomAnnotationPrint(Annotation annotation)
{
Console.WriteLine("-------------");
Console.WriteLine("Custom print:");
Console.WriteLine("-------------");
var sentences = (ArrayList)annotation.get(new CoreAnnotations.SentencesAnnotation().getClass());
foreach (CoreMap sentence in sentences)
{
Console.WriteLine("\n\nSentence : '{0}'", sentence);
var tokens = (ArrayList)sentence.get(new CoreAnnotations.TokensAnnotation().getClass());
foreach (CoreLabel token in tokens)
{
var word = token.get(new CoreAnnotations.TextAnnotation().getClass());
var pos = token.get(new CoreAnnotations.PartOfSpeechAnnotation().getClass());
var ner = token.get(new CoreAnnotations.NamedEntityTagAnnotation().getClass());
Console.WriteLine("{0} \t[pos={1}; ner={2}]", word, pos, ner);
}
Console.WriteLine("\nTree:");
var tree = (Tree)sentence.get(new TreeCoreAnnotations.TreeAnnotation().getClass());
using (var stream = new ByteArrayOutputStream())
{
tree.pennPrint(new PrintWriter(stream));
Console.WriteLine("The first sentence parsed is:\n {0}", stream.toString());
}
Console.WriteLine("\nDependencies:");
var deps = (SemanticGraph)sentence.get(new SemanticGraphCoreAnnotations.CollapsedDependenciesAnnotation().getClass());
foreach (SemanticGraphEdge edge in deps.edgeListSorted().toArray())
{
var gov = edge.getGovernor();
var dep = edge.getDependent();
Console.WriteLine(
"{0}({1}-{2},{3}-{4})", edge.getRelation(),
gov.word(), gov.index(), dep.word(), dep.index());
}
}
}
public void ProcessText(string text, string outputPath)
{
var props = new Properties();
props.setProperty("annotators", "tokenize, ssplit, pos, lemma, ner, parse");
props.setProperty("ner.useSUTime", "0");
// We should change current directory, so StanfordCoreNLP could find all the model files automatically
var curDir = Environment.CurrentDirectory;
Directory.SetCurrentDirectory(stanfordJarRoot);
var pipeline = new StanfordCoreNLP(props);
Directory.SetCurrentDirectory(curDir);
Console.WriteLine("Starting to parse.");
// Annotation
var annotation = new Annotation(text);
pipeline.annotate(annotation);
// Result - Pretty Print
Console.WriteLine("Parsing complete.. writing to file.");
string jsonOutput;
using (var stream = new ByteArrayOutputStream())
{
pipeline.jsonPrint(annotation, new PrintWriter(stream));
jsonOutput = stream.toString();
stream.close();
}
using (var file = new StreamWriter(outputPath))
{
file.WriteLine(jsonOutput);
}
Console.WriteLine("Processing complete.");
}
public static void Main()
{
// Path to the folder with models extracted from `stanford-corenlp-3.6.0-models.jar`
var jarRoot = @"data";
// Text for processing
var text = "Kosgi Santosh sent an email to Stanford University. He didn't get a reply.";
// var text = "The following transaction was received from Admin entered on 04/30/2016 at 11:31 PM CDT and filed on 04/28/2016 ";
// Annotation pipeline configuration
var props = new Properties();
props.setProperty("annotators", "tokenize, ssplit, pos, lemma, ner, parse, dcoref");
props.setProperty("ner.useSUTime", "0");
// We should change current directory, so StanfordCoreNLP could find all the model files automatically
var curDir = Environment.CurrentDirectory;
Directory.SetCurrentDirectory(jarRoot);
var pipeline = new StanfordCoreNLP(props);
Directory.SetCurrentDirectory(curDir);
// Annotation
var annotation = new Annotation(text);
pipeline.annotate(annotation);
// Result - Pretty Print
using (var stream = new ByteArrayOutputStream())
{
pipeline.prettyPrint(annotation, new PrintWriter(stream));
Console.WriteLine(stream.toString());
stream.close();
}
}
static void Main(string[] args)
{
var jarRoot = @"..\..\..\es\";
/* CultureInfo ci = new CultureInfo("es-ES");
* Thread.CurrentThread.CurrentCulture = ci;
* Thread.CurrentThread.CurrentUICulture = ci;*/
// Text for processing
var text = "\"O pior erro que poderíamos cometer era, à boleia do crescimento económico, termos a ilusão de que os problemas estruturais da zona euro ficaram resolvidos\", defende António Costa, que deu uma entrevista ao Público que será publicada este domingo e onde o primeiro-ministro diz que a moeda única foi um bonus para a economia alemã.";
string sb = "António Costa defende que, na criação da moeda única, houve um \"excesso de voluntarismo político\" e nem todos terão percebido que \"o euro foi o maior bónus à competitividade da economia alemã que a Europa lhe poderia ter oferecido\".";
var propsFile = Path.Combine(jarRoot, "StanfordCoreNLP-spanish.properties");
// Annotation pipeline configuration
var props = new Properties();
// props.setProperty("annotators", "tokenize, ssplit, pos, lemma, ner, depparse, kbp");
props.load(new FileReader(propsFile));
props.put("ner.useSUTime", "0");
props.put("tokenize.verbose", "true");
// props.setProperty("annotators", "tokenize, ssplit, pos, lemma, ner, depparse, kbp, coref,entitymentions,quote1");
props.setProperty("annotators", "tokenize, ssplit, pos, lemma, ner, depparse, kbp, coref,entitymentions, quote");
//props.setProperty("customAnnotatorClass.quote1", "edu.stanford.nlp.pipeline.QuoteAnnotator");
//props.setProperty("quote1.attributeQuotes", "false");
/*String modPath = @"C:\Users\anacorreia\source\repos\ConsoleApp1\es\";
* props.put("pos.model", modPath + @"edu\stanford\nlp\models\pos-tagger\spanish\spanish-ud.tagger");
* props.put("tokenize.language","es");
*
* props.put("ner.model", modPath + @"edu\stanford\nlp\models\ner\spanish.ancora.distsim.s512.crf.ser.gz");
* props.put("ner.applyNumericClassifiers", "1");
*
* props.put("ner.language", "1");
* props.put("sutime.language", "1");
*
* props.put("parse.model", modPath + @"edu\stanford\nlp\models\lexparser\spanishPCFG.ser.gz");
* props.put("depparse.model", modPath + @"edu\stanford\nlp\models\parser\nndep\UD_Spanish.gz");
* props.put("depparse.language", "spanish");
*
* props.put("ner.fine.regexner.mapping", modPath + @"edu\stanford\nlp\models\kbp\spanish\gazetteers\");
* props.put("ner.fine.regexner.validpospattern", "^(NOUN|ADJ|PROPN).*");
* props.put("ner.fine.regexner.ignorecase", "1");
* props.put("ner.fine.regexner.noDefaultOverwriteLabels", "CITY,COUNTRY,STATE_OR_PROVINCE");
*
* props.put("kbp.semgrex", modPath + @"edu\stanford\nlp\models\kbp\spanish\semgrex");
* props.put("kbp.tokensregex", modPath + @"edu\stanford\nlp\models\kbp\spanish\tokensregex");
* props.put("kbp.model", "none");
* props.put("kbp.language", "es");
*
* props.put("entitylink.caseless", "1");
* props.put("entitylink.wikidict", modPath + @"edu\stanford\nlp\models\kbp\spanish\wikidict_spanish.tsv");
*/
// We should change current directory, so StanfordCoreNLP could find all the model files automatically
var curDir = Environment.CurrentDirectory;
Directory.SetCurrentDirectory(jarRoot);
var pipeline = new StanfordCoreNLP(props);
// Annotation
CoreDocument testDoc = new CoreDocument(sb);
pipeline.annotate(testDoc);
// var annotation = new Annotation(sb);
//pipeline.annotate(annotation);
string sxc = "\"O pior erro que poderíamos cometer era, à boleia do crescimento económico, termos a ilusão de que os problemas estruturais da zona euro ficaram resolvidos\", defende António Costa, que deu uma entrevista ao Público que será publicada este domingo e onde o primeiro-ministro diz que a moeda única foi um \"bónus\" para a economia alemã. Numa entrevista em que, a julgar pelo excerto publicado para já, se focou essencialmente em questões europeias, e não na política interna, o primeiro-ministro mostrou-se, também, favorável à introdução de impostos europeus." +
"António Costa defende que, na criação da moeda única, houve um \"excesso de voluntarismo político\" e nem todos terão percebido que \"o euro foi o maior bónus à competitividade da economia alemã que a Europa lhe poderia ter oferecido\". Agora, a menos que se tomem medidas de correção das assimetrias, \"a zona euro será mais uma vez confrontada com uma crise como a que vivemos agora\"." +
"O objetivo de todos os líderes europeus deve ser evitar que se volte a cometer \"o erro que nos acompanhou desde 2000 até 2011\", isto é, marcar passo nas reformas. E uma das principais ferramentas de que a zona euro necessita é de um orçamento da zona euro destinado a financiar reformas para acelerar a convergência das economias." +
"Com a saída do Reino Unido e com a necessidade de investir na segurança, na defesa e na ciência, António Costa defende que \"ou estamos disponíveis para sacrificar a parte do Orçamento afeta às políticas de coesão e à PAC, ou temos de encontrar outras fontes de receita\". Onde? Mais contribuições dos Estados, isto é, \"mais impostos dos portugueses\", ou receitas próprias criadas pela União, nomeadamente através de impostos europeus.";
CoreDocument testDoc3 = new CoreDocument(sxc);
pipeline.annotate(testDoc3);
// Result - Pretty Print
using (var stream = new ByteArrayOutputStream())
{
pipeline.prettyPrint(testDoc.annotation(), new PrintWriter(stream));
Console.WriteLine("Stream testDoc3 " + stream.toString());
stream.close();
}
using (var stream = new ByteArrayOutputStream())
{
pipeline.prettyPrint(testDoc3.annotation(), new PrintWriter(stream));
Console.WriteLine("Stream testDoc3 " + stream.toString());
stream.close();
}
Directory.SetCurrentDirectory(curDir);
}
private object StanfordNLP(string input)
{
string npath = Directory.GetCurrentDirectory();
string NLPquery = "";
string getNoun = "";
bool compound = false;
var text = input;
// Annotation pipeline configuration
var props = new Properties();
props.setProperty("annotators", "tokenize, ssplit, pos, lemma, ner, parse, dcoref");
props.setProperty("pos.model", npath + @"\edu\stanford\nlp\models\pos-tagger\english-bidirectional\english-bidirectional-distsim.tagger");
props.setProperty("ner.model", npath + @"\edu\stanford\nlp\models\ner\english.all.3class.distsim.crf.ser.gz");
props.setProperty("parse.model", npath + @"\edu\stanford\nlp\models\lexparser\englishPCFG.ser.gz");
props.setProperty("dcoref.demonym", npath + @"\edu\stanford\nlp\models\dcoref\demonyms.txt");
props.setProperty("dcoref.states", npath + @"\edu\stanford\nlp\models\dcoref\state-abbreviations.txt");
props.setProperty("dcoref.animate", npath + @"\edu\stanford\nlp\models\dcoref\animate.unigrams.txt");
props.setProperty("dcoref.inanimate", npath + @"\edu\stanford\nlp\models\dcoref\inanimate.unigrams.txt");
props.setProperty("dcoref.male", npath + @"\edu\stanford\nlp\models\dcoref\male.unigrams.txt");
props.setProperty("dcoref.neutral", npath + @"\edu\stanford\nlp\models\dcoref\neutral.unigrams.txt");
props.setProperty("dcoref.female", npath + @"\edu\stanford\nlp\models\dcoref\female.unigrams.txt");
props.setProperty("dcoref.plural", npath + @"\edu\stanford\nlp\models\dcoref\plural.unigrams.txt");
props.setProperty("dcoref.singular", npath + @"\edu\stanford\nlp\models\dcoref\singular.unigrams.txt");
props.setProperty("dcoref.countries", npath + @"\edu\stanford\nlp\models\dcoref\countries");
props.setProperty("dcoref.extra.gender", npath + @"\edu\stanford\nlp\models\dcoref\namegender.combine.txt");
props.setProperty("dcoref.states.provinces", npath + @"\edu\stanford\nlp\models\dcoref\statesandprovinces");
props.setProperty("dcoref.singleton.predictor", npath + @"\edu\stanford\nlp\models\dcoref\singleton.predictor.ser");
props.setProperty("dcoref.big.gender.number", npath + @"\edu\stanford\nlp\models\dcoref\gender.map.ser.gz");
props.setProperty("sutime.rules", npath + @"\edu\stanford\nlp\models\sutime\defs.sutime.txt, " + npath + @"\edu\stanford\nlp\models\sutime\english.holidays.sutime.txt, " + npath + @"\edu\stanford\nlp\models\sutime\english.sutime.txt");
props.setProperty("sutime.binders", "0");
props.setProperty("ner.useSUTime", "0");
var pipeline = new StanfordCoreNLP(props);
// Annotation
var annotation = new Annotation(text);
pipeline.annotate(annotation);
List <string> NLPDATA = new List <string>();
using (var stream = new ByteArrayOutputStream())
{
// pipeline.prettyPrint(annotation, new PrintWriter(stream));
pipeline.conllPrint(annotation, new PrintWriter(stream));
string output = stream.toString();
Console.WriteLine(output);
string[] lines = Regex.Split(output, "[\r\n]+");
// Console.WriteLine(lines.Length);
string[][] wordMatrix = new string[lines.Length][];
for (var i = 0; i < wordMatrix.Length; i++)
{
wordMatrix[i] = new string[10];
string[] words = Regex.Split(lines[i], "[^a-zA-Z0-9]+");
// Console.WriteLine(words.Length);
for (int ii = 0; ii < words.Length; ii++)
{
wordMatrix[i][ii] = words[ii];
}
}
for (int i = 0; i < lines.Length; i++)
{
for (int ii = 0; ii < wordMatrix[i].Length; ii++)
{
if (wordMatrix[i][ii] == "VB" || wordMatrix[i][ii] == "RP" || wordMatrix[i][ii] == "NN" || wordMatrix[i][ii] == "NNP")
{
NLPDATA.Add(wordMatrix[i][ii] + " " + wordMatrix[i][ii - 1] + " " + wordMatrix[i][6]);
NLPquery = NLPquery + " " + wordMatrix[i][ii - 1];
}
if (wordMatrix[i][ii] == "NN" || wordMatrix[i][ii] == "NNP" || wordMatrix[i][ii] == "NNS")
{
if (wordMatrix[i][6] == "compound" || wordMatrix[i][6] == "xcomp")
{
getNoun = wordMatrix[i][ii - 1];
compound = true;
}
if (wordMatrix[i][6] == "dep" && compound != true && wordMatrix[i][ii] == "NN")
{
getNoun = wordMatrix[i][ii - 1];
}
if (wordMatrix[i][6] == "dobj" && compound != true && wordMatrix[i][ii] == "NN")
{
getNoun = wordMatrix[i][ii - 1];
}
}
}
}
stream.close();
}
//Intent, Name,
NLP nLP = new NLP();
// nLP.getAnswer(NLPquery.Trim());
// Console.WriteLine(nLP.getProbability(getNoun.Trim()).ToString());
string action = nLP.getAnswer(NLPquery.Trim());
string nprob = nLP.getProbability(getNoun.Trim()).ToString();
string aprob = nLP.getProbability(NLPquery.Trim()).ToString();
if (action == getNoun)
{
nprob = "0";
aprob = "0";
}
compound = false;
return(new { intent = action, noun = getNoun, action_prob = aprob, noun_prob = nprob });
}