fixed multi term of standford ner
[pnews.git] / war / src / main / java / pnews / NER.java
1 package pnews;
2
3 import java.io.IOException;
4 import java.util.ArrayList;
5 import java.util.List;
6 import java.util.logging.Logger;
7
8 import edu.stanford.nlp.ie.crf.CRFClassifier;
9 import edu.stanford.nlp.ling.CoreLabel;
10 import edu.stanford.nlp.util.Triple;
11
12 /** https://stanfordnlp.github.io/CoreNLP/api.html */
13 public class NER {
14         private static final String CLASS_NAME = NER.class.getName();
15         private static final Logger LOG = Logger.getLogger(CLASS_NAME); 
16         
17         public static List<String> classify(String str, List<String> entities) throws ClassCastException, ClassNotFoundException, IOException {
18                 final CRFClassifier<CoreLabel> classifier = CRFClassifier.getDefaultClassifier();
19                 List<Triple<String, Integer, Integer>> triples;
20                 String w;
21                 final String FUNCTION_NAME = "classify";                       
22                 
23                 LOG.entering(CLASS_NAME, FUNCTION_NAME, str);
24
25                 OpenNLP.classify(str, entities);
26                                 
27                 synchronized (classifier) {
28                         triples = classifier.classifyToCharacterOffsets(str);
29                         for (Triple<String, Integer, Integer> t: triples) {
30                                 w = str.substring(t.second, t.third);
31                                 if (!entities.contains(w))
32                                         entities.add(w);
33                         }
34                 }
35                 
36                 entities.remove("CNET");
37                 entities.remove("Read More");
38                 entities.remove("New");
39                 entities.remove("App");
40                 
41                 LOG.exiting(CLASS_NAME, FUNCTION_NAME, entities);
42                 
43                 return entities;
44         }
45         
46         public static void main(String[] args) throws Exception {
47                 List<String> lst;
48                 
49                 lst = classify("I live in Washington and New York in United States.", new ArrayList<>());
50                 for (String str: lst)
51                         System.out.println(str);
52         }
53 }