X-Git-Url: http://git.wpitchoune.net/gitweb/?a=blobdiff_plain;f=war%2Fsrc%2Fmain%2Fjava%2Fpnews%2FNER.java;h=5e7ce292a143a3865dd94ff7e143bb8028798740;hb=88a7ba9745b8318ca6c4f741906a40e3d6a8f07e;hp=ac34c0812faf0f24f71392b497ac6fc86ea5d608;hpb=c6f5722d9572158f0d561819f080f8ccf4e3c7d9;p=pnews.git diff --git a/war/src/main/java/pnews/NER.java b/war/src/main/java/pnews/NER.java index ac34c08..5e7ce29 100644 --- a/war/src/main/java/pnews/NER.java +++ b/war/src/main/java/pnews/NER.java @@ -6,39 +6,35 @@ import java.util.List; import java.util.logging.Logger; import edu.stanford.nlp.ie.crf.CRFClassifier; -import edu.stanford.nlp.ling.CoreAnnotations.AnswerAnnotation; -import edu.stanford.nlp.ling.CoreLabel; +import edu.stanford.nlp.util.CoreMap; +import edu.stanford.nlp.util.Triple; +import pnews.servlet.Config; /** https://stanfordnlp.github.io/CoreNLP/api.html */ public class NER { private static final String CLASS_NAME = NER.class.getName(); - private static final Logger LOG = Logger.getLogger(CLASS_NAME); + private static final Logger LOG = Logger.getLogger(CLASS_NAME); + private static final CRFClassifier classifier = CRFClassifier.getDefaultClassifier(); - public static List classify(String str, List entities) throws ClassCastException, ClassNotFoundException, IOException { - CRFClassifier classifier; - List> out; - String cat, w; - final String FUNCTION_NAME = "classify"; + public static List classify(String str, List entities, Config config) throws ClassCastException, ClassNotFoundException, IOException { + + List> triples; + String w; + final String FUNCTION_NAME = "classify"; LOG.entering(CLASS_NAME, FUNCTION_NAME, str); - OpenNLP.classify(str, entities); - - classifier = CRFClassifier.getDefaultClassifier(); - out = classifier.classify(str); - - for (List labels: out) - for (CoreLabel l: labels) { - cat = l.getString(AnswerAnnotation.class); - w = l.word(); - if (!cat.equals("O") && !entities.contains(w)) - entities.add(w); - } - - entities.remove("CNET"); - entities.remove("Read More"); - entities.remove("New"); - entities.remove("App"); + OpenNLP.classify(str, entities, config); + + synchronized (classifier) { + triples = classifier.classifyToCharacterOffsets(str); + } + + for (Triple t: triples) { + w = str.substring(t.second, t.third); + if (!config.isBlacklistedEntity(w) && !entities.contains(w)) + entities.add(config.getEntityAlias(w)); + } LOG.exiting(CLASS_NAME, FUNCTION_NAME, entities); @@ -46,6 +42,10 @@ public class NER { } public static void main(String[] args) throws Exception { - classify("I live in Washington.", new ArrayList<>()); + List lst; + + lst = classify("I live in Washington and New York in United States.", new ArrayList<>(), new Config()); + for (String str: lst) + System.out.println(str); } } \ No newline at end of file