@Override
protected List<Keyterm> getKeyterms(String question) {
String document = question;
System.out.println("CHECK " + question);
// int indexSpace = documentLine.indexOf(' ');
// String docID = documentLine.substring(0, indexSpace);
// String document = documentLine.substring(indexSpace + 1);
Chunker extractor = null;
try {
extractor = (Chunker) AbstractExternalizable.readObject(modelFile);
} catch (IOException e) {
// TODO Auto-generated catch block
e.printStackTrace();
} catch (ClassNotFoundException e) {
// TODO Auto-generated catch block
e.printStackTrace();
}
List<Keyterm> keyterms = new ArrayList<Keyterm>();
Chunking entities = extractor.chunk(document);
for (Chunk ne : entities.chunkSet()) {
// Annotations
String keyWord = question.substring(ne.start(), ne.end());
keyterms.add(new Keyterm(keyWord));
}
// store the keyterms from chunker here from annotator
return keyterms;
}
public static void main(String[] args) {
String modelFile = "src/main/resources/models/ne-en-bio-genetag.HmmChunker";
String question =
"What is the role of PrnP in mad cow disease (bovine spongiform encephalitis, BSE)?";
try {
Chunker achunker = (Chunker) AbstractExternalizable.readObject(new File(modelFile));
Chunking chunking = achunker.chunk(question);
for (Chunk chunk : chunking.chunkSet()) {
System.out.println(question.substring(chunk.start(), chunk.end()));
}
} catch (Exception e) {
System.err.println("oops");
}
}
@Override
protected List<Keyterm> getKeyterms(String text) {
List<Keyterm> keyTerms = new ArrayList<Keyterm>();
Chunking chunking = chunker.chunk(text);
for (Chunk chunk : chunking.chunkSet()) {
int start = chunk.start();
int end = chunk.end();
String term = text.substring(start, end);
keyTerms.add(new Keyterm(term));
}
String sentenceText = text;
for (int start = 0; start < sentenceText.length(); start++) {
if (start != 0 && sentenceText.charAt(start - 1) != ' ') continue;
for (int end = start + 1; end < sentenceText.length() && end < start + 97; end++) {
if (end != sentenceText.length() - 1 && sentenceText.charAt(end) != ' ') continue;
String spanText = sentenceText.substring(start, end);
if (geneDictionary.contains(spanText)) {
keyTerms.add(new Keyterm(spanText));
}
}
}
return keyTerms;
}
@Override
protected List<Keyterm> updateKeyterms(String question, List<Keyterm> keyterms) {
Chunking chunking = chunker.chunk(question);
for (Chunk chunk : chunking.chunkSet()) {
String word = question.substring(chunk.start(), chunk.end());
Keyterm keyterm = new Keyterm(word);
keyterm.setComponentId("GENE");
keyterms.add(keyterm);
}
return keyterms;
}
void assertChunkingResult(Chunker chunker, String in, String[] types, int[] starts, int[] ends)
throws IOException {
Chunking chunking = chunker.chunk(in);
Chunk[] chunks = chunking.chunkSet().toArray(new Chunk[0]);
Arrays.sort(chunks, Chunk.TEXT_ORDER_COMPARATOR);
assertEquals(types.length, starts.length);
assertEquals(starts.length, ends.length);
assertEquals(chunks.length, types.length);
for (int i = 0; i < chunks.length; ++i) {
assertEquals(types[i], chunks[i].type());
assertEquals(starts[i], chunks[i].start());
assertEquals(ends[i], chunks[i].end());
}
}
/** annotate on the cas where we have gene names */
public void process(JCas aJCas) {
if (chunker == null) {
initialize();
}
// get document text
String docText = aJCas.getDocumentText();
Chunking chunking = chunker.chunk(docText);
for (Chunk ck : chunking.chunkSet()) {
GenTag annotation = new GenTag(aJCas);
annotation.setBegin(ck.start());
annotation.setEnd(ck.end());
annotation.addToIndexes();
}
}