import java.util.logging.Logger;
import edu.stanford.nlp.ie.crf.CRFClassifier;
-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 ThreadLocal<CRFClassifier<CoreLabel>> classifier = new ThreadLocal<CRFClassifier<CoreLabel>>() {
- @Override
- protected CRFClassifier<CoreLabel> initialValue() {
- return CRFClassifier.getDefaultClassifier();
- }
- };
+ private static final CRFClassifier<CoreMap> classifier = CRFClassifier.getDefaultClassifier();
- public static List<String> classify(String str, List<String> entities) throws ClassCastException, ClassNotFoundException, IOException {
+ public static List<String> classify(String str, List<String> entities, Config config) throws ClassCastException, ClassNotFoundException, IOException {
List<Triple<String, Integer, Integer>> triples;
String w;
LOG.entering(CLASS_NAME, FUNCTION_NAME, str);
- OpenNLP.classify(str, entities);
+ OpenNLP.classify(str, entities, config);
synchronized (classifier) {
- triples = classifier.get().classifyToCharacterOffsets(str);
- for (Triple<String, Integer, Integer> t: triples) {
- w = str.substring(t.second, t.third);
- if (!entities.contains(w))
- entities.add(w);
- }
+ triples = classifier.classifyToCharacterOffsets(str);
+ }
+
+ for (Triple<String, Integer, Integer> t: triples) {
+ w = str.substring(t.second, t.third);
+ if (!config.isBlacklistedEntity(w) && !entities.contains(w))
+ entities.add(config.getEntityAlias(w));
}
-
- entities.remove("CNET");
- entities.remove("Read More");
- entities.remove("New");
- entities.remove("App");
LOG.exiting(CLASS_NAME, FUNCTION_NAME, entities);
public static void main(String[] args) throws Exception {
List<String> lst;
- lst = classify("I live in Washington and New York in United States.", new ArrayList<>());
+ lst = classify("I live in Washington and New York in United States.", new ArrayList<>(), new Config());
for (String str: lst)
System.out.println(str);
}