Package org.ictclas4j.bean

Examples of org.ictclas4j.bean.POS

157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
          for (int j = 0; wis != null && j < wis.size(); j++) {
            WordItem wi = wis.get(j);
            int tag = wi.getHandle();
            double freq = -Math.log((1 + wi.getFreq()));
            freq += Math.log((context.getFreq(0, wi.getHandle()) + wis.size() + 1));
            POS pos = new POS(tag, freq);
            sn.addPos(pos);
          }

          if (Utility.SENTENCE_BEGIN.equals(curWord))
            sn.addPos(new POS(100, 0));

          else if (Utility.SENTENCE_END.equals(curWord))
            sn.addPos(new POS(101, 0));
          else {
            int nFreq = 0;
            wis = coreDict.getHandle(curWord);
            if (wis != null) {
              for (WordItem wi : wis)
                nFreq += wi.getFreq();

              if (wis.size() > 0) {
                double freq = -Math.log((double) (1 + nFreq));
                freq += Math.log((double) (context.getFreq(0, 0) + wis.size()));
                sn.addPos(new POS(0, freq));
              }
            }
          }
        } else {
          if (sn.getPos() > 0) {
            int tag = sn.getPos();
            double value = -Math.log(sn.getValue());
            value += Math.log(context.getFreq(0, tag));
            if (value < 0)
              value = 0;
            sn.addPos(new POS(tag, value));
          } else {
            if (sn.getPos() < 0) {
              sn.setPos(-sn.getPos());
              sn.addPos(new POS(-sn.getPos(), sn.getValue()));
            }
            ArrayList<WordItem> wis = coreDict.getHandle(curWord);
            if (wis != null) {
              for (WordItem wi : wis) {
                int tag = wi.getHandle();
                double value = -Math.log(1 + wi.getFreq());
                value += Math.log(context.getFreq(0, tag) + wis.size());
                sn.addPos(new POS(tag, value));
              }
            }
          }
        }

        if (sn.getAllPos() == null)
          guessPos(tagType, sn);
       
        // ���һ���ʽڵ��Ӧ��allPosΪnull����˵�����޷������ɴ�
        // ���Ĵ�������һ���ʵĴ���
        if (i - 1 >= 0 && sns.get(i - 1).getPosSize() == -1) {
          if (sn.getPosSize() > 0) {
            POS pos = new POS(sn.getAllPos().get(0).getTag(), 0);
            sns.get(i - 1).addPos(pos);
          }
        }
      }

      // ���һ��������
      SegNode last = sns.get(i - 1);
      if (last != null) {
        SegNode sn = new SegNode();
        int tag = 0;
        if (tagType != Utility.TAG_TYPE.TT_NORMAL)
          tag = 101;
        else
          tag = 1;
        POS pos = new POS(tag, 0);
        sn.addPos(pos);
        sns.add(sn);
      }
    }
  }
View Full Code Here
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
    if (sns != null && context != null) {
      for (int i = 0; i < sns.size(); i++) {
        if (i == 0) {
          int pos = tagType != Utility.TAG_TYPE.TT_NORMAL ? 100 : 0;
          prevAllPos = new ArrayList<POS>();
          prevAllPos.add(new POS(pos, 0));
        } else {
          prevAllPos = sns.get(i - 1).getAllPos();
        }
        allPos = sns.get(i).getAllPos();
        for (POS pos : allPos) {
          int minPrev = 0;
          double minFreq = 1000;
          for (int k = 0; k < prevAllPos.size(); k++) {
            POS prevPos = prevAllPos.get(k);
            double temp = context.getPossibility(0, prevPos.getTag(), pos.getTag());
            temp = -Math.log(temp) + prevPos.getFreq();
            if (temp < minFreq) {
              minFreq = temp;
              minPrev = k;
            }
          }
View Full Code Here
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
      case TT_NORMAL:
        break;
      case TT_PERSON:
        if (word.indexOf("����") != -1) {
          freq = (double) 1 / (double) (context.getFreq(0, 6) + 1);
          sn.addPos(new POS(6, freq));
        } else {
          freq = (double) 1 / (double) (context.getFreq(0, 0) + 1);
          sn.addPos(new POS(0, freq));

          if (sn.getLen() >= 4) {
            freq = (double) 1 / (double) (context.getFreq(0, 0) + 1);
            sn.addPos(new POS(0, freq));
            freq = (double) 1 / (double) (context.getFreq(0, 11) * 8);
            sn.addPos(new POS(11, freq));
            freq = (double) 1 / (double) (context.getFreq(0, 12) * 8);
            sn.addPos(new POS(12, freq));
            freq = (double) 1 / (double) (context.getFreq(0, 13) * 8);
            sn.addPos(new POS(13, freq));
          } else if (sn.getLen() == 2) {
            freq = (double) 1 / (double) (context.getFreq(0, 0) + 1);
            sn.addPos(new POS(0, freq));
            charType = Utility.charType(word);
            if (charType == Utility.CT_OTHER || charType == Utility.CT_CHINESE) {
              freq = (double) 1 / (double) (context.getFreq(0, 1) + 1);
              sn.addPos(new POS(1, freq));
              freq = (double) 1 / (double) (context.getFreq(0, 2) + 1);
              sn.addPos(new POS(2, freq));
              freq = (double) 1 / (double) (context.getFreq(0, 3) + 1);
              sn.addPos(new POS(3, freq));
              freq = (double) 1 / (double) (context.getFreq(0, 4) + 1);
              sn.addPos(new POS(4, freq));
            }
            freq = (double) 1 / (double) (context.getFreq(0, 11) * 8);
            sn.addPos(new POS(11, freq));
            freq = (double) 1 / (double) (context.getFreq(0, 12) * 8);
            sn.addPos(new POS(12, freq));
            freq = (double) 1 / (double) (context.getFreq(0, 13) * 8);
            sn.addPos(new POS(13, freq));
          }
        }
        break;
      case TT_PLACE:
        freq = (double) 1 / (double) (context.getFreq(0, 0) + 1);
        sn.addPos(new POS(0, freq));

        if (sn.getLen() >= 4) {
          freq = (double) 1 / (double) (context.getFreq(0, 11) * 8);
          sn.addPos(new POS(11, freq));
          freq = (double) 1 / (double) (context.getFreq(0, 12) * 8);
          sn.addPos(new POS(12, freq));
          freq = (double) 1 / (double) (context.getFreq(0, 13) * 8);
          sn.addPos(new POS(13, freq));
        } else if (sn.getLen() == 2) {
          freq = (double) 1 / (double) (context.getFreq(0, 0) + 1);
          sn.addPos(new POS(0, freq));
          charType = Utility.charType(word);
          if (charType == Utility.CT_OTHER || charType == Utility.CT_CHINESE) {

            freq = (double) 1 / (double) (context.getFreq(0, 1) + 1);
            sn.addPos(new POS(1, freq));
            freq = (double) 1 / (double) (context.getFreq(0, 2) + 1);
            sn.addPos(new POS(2, freq));
            freq = (double) 1 / (double) (context.getFreq(0, 3) + 1);
            sn.addPos(new POS(3, freq));
            freq = (double) 1 / (double) (context.getFreq(0, 4) + 1);
            sn.addPos(new POS(4, freq));
          }
          freq = (double) 1 / (double) (context.getFreq(0, 11) * 8);
          sn.addPos(new POS(11, freq));
          freq = (double) 1 / (double) (context.getFreq(0, 12) * 8);
          sn.addPos(new POS(12, freq));
          freq = (double) 1 / (double) (context.getFreq(0, 13) * 8);
          sn.addPos(new POS(13, freq));
        }
        break;
      case TT_TRANS_PERSON:
        freq = (double) 1 / (double) (context.getFreq(0, 0) + 1);
        sn.addPos(new POS(0, freq));
        if (!Utility.isAllChinese(word)) {
          if (Utility.isAllLetter(word)) {
            freq = (double) 1 / (double) (context.getFreq(0, 1) + 1);
            sn.addPos(new POS(1, freq));
            freq = (double) 1 / (double) (context.getFreq(0, 11) + 1);
            sn.addPos(new POS(11, freq));
            freq = (double) 1 / (double) (context.getFreq(0, 2) * 2 + 1);
            sn.addPos(new POS(2, freq));
            freq = (double) 1 / (double) (context.getFreq(0, 3) * 2 + 1);
            sn.addPos(new POS(3, freq));
            freq = (double) 1 / (double) (context.getFreq(0, 12) * 2 + 1);
            sn.addPos(new POS(12, freq));
            freq = (double) 1 / (double) (context.getFreq(0, 13) * 2 + 1);
            sn.addPos(new POS(13, freq));
          }
          freq = (double) 1 / (double) (context.getFreq(0, 41) * 8);
          sn.addPos(new POS(41, freq));
          freq = (double) 1 / (double) (context.getFreq(0, 42) * 8);
          sn.addPos(new POS(42, freq));
          freq = (double) 1 / (double) (context.getFreq(0, 43) * 8);
          sn.addPos(new POS(43, freq));
        } else if (sn.getLen() >= 4) {
          freq = (double) 1 / (double) (context.getFreq(0, 41) * 8);
          sn.addPos(new POS(41, freq));
          freq = (double) 1 / (double) (context.getFreq(0, 42) * 8);
          sn.addPos(new POS(42, freq));
          freq = (double) 1 / (double) (context.getFreq(0, 43) * 8);
          sn.addPos(new POS(43, freq));
        } else if (sn.getLen() == 2) {
          charType = Utility.charType(word);
          if (charType == Utility.CT_OTHER || charType == Utility.CT_CHINESE) {
            freq = (double) 1 / (double) (context.getFreq(0, 1) * 2 + 1);
            sn.addPos(new POS(1, freq));
            freq = (double) 1 / (double) (context.getFreq(0, 2) * 2 + 1);
            sn.addPos(new POS(2, freq));
            freq = (double) 1 / (double) (context.getFreq(0, 3) * 2 + 1);
            sn.addPos(new POS(3, freq));
            freq = (double) 1 / (double) (context.getFreq(0, 30) * 8 + 1);
            sn.addPos(new POS(30, freq));
            freq = (double) 1 / (double) (context.getFreq(0, 11) * 4 + 1);
            sn.addPos(new POS(11, freq));
            freq = (double) 1 / (double) (context.getFreq(0, 12) * 4 + 1);
            sn.addPos(new POS(12, freq));
            freq = (double) 1 / (double) (context.getFreq(0, 13) * 4 + 1);
            sn.addPos(new POS(13, freq));
            freq = (double) 1 / (double) (context.getFreq(0, 21) * 2 + 1);
            sn.addPos(new POS(21, freq));
            freq = (double) 1 / (double) (context.getFreq(0, 22) * 2 + 1);
            sn.addPos(new POS(22, freq));
            freq = (double) 1 / (double) (context.getFreq(0, 23) * 2 + 1);
            sn.addPos(new POS(23, freq));
          }
          freq = (double) 1 / (double) (context.getFreq(0, 41) * 8);
          sn.addPos(new POS(41, freq));
          freq = (double) 1 / (double) (context.getFreq(0, 42) * 8);
          sn.addPos(new POS(42, freq));
          freq = (double) 1 / (double) (context.getFreq(0, 43) * 8);
          sn.addPos(new POS(43, freq));
        }
        break;
      default:
        break;
      }
View Full Code Here
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
      // �����ǿ�ʼ�ͽ������
      for (int i = size - 1, j = 0; i >= 0; i--) {
        ArrayList<POS> allPos = sns.get(i).getAllPos();
        if (allPos != null && allPos.size() > j) {
          POS pos = allPos.get(j);
          pos.setBest(true);
          j = pos.getPrev();
        } else if (i + 1 < size - 1) {
          int tag = getBestTag(sns.get(i + 1));
          POS pos = new POS(tag, 0);
          pos.setBest(true);
          sns.get(i).addPos(pos);
        }
      }
      // �ѽ�����ȥ�����õ�����Ŀ�Ľ�����Ϊ�˵õ����һ����ĩ����ĩ���ʵ����Ŵ���
View Full Code Here
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
          for (int j = 0; wis != null && j < wis.size(); j++) {
            WordItem wi = wis.get(j);
            int tag = wi.getHandle();
            double freq = -Math.log((1 + wi.getFreq()));
            freq += Math.log((context.getFreq(0, wi.getHandle()) + wis.size() + 1));
            POS pos = new POS(tag, freq);
            sn.addPos(pos);
          }

          if (Utility.SENTENCE_BEGIN.equals(curWord))
            sn.addPos(new POS(100, 0));

          else if (Utility.SENTENCE_END.equals(curWord))
            sn.addPos(new POS(101, 0));
          else {
            int nFreq = 0;
            wis = coreDict.getHandle(curWord);
            if (wis != null) {
              for (WordItem wi : wis)
                nFreq += wi.getFreq();

              if (wis.size() > 0) {
                double freq = -Math.log((double) (1 + nFreq));
                freq += Math.log((double) (context.getFreq(0, 0) + wis.size()));
                sn.addPos(new POS(0, freq));
              }
            }
          }
        } else {
          if (sn.getPos() > 0) {
            int tag = sn.getPos();
            double value = -Math.log(sn.getValue());
            value += Math.log(context.getFreq(0, tag));
            if (value < 0)
              value = 0;
            sn.addPos(new POS(tag, value));
          } else {
            if (sn.getPos() < 0) {
              sn.setPos(-sn.getPos());
              sn.addPos(new POS(-sn.getPos(), sn.getValue()));
            }
            ArrayList<WordItem> wis = coreDict.getHandle(curWord);
            if (wis != null) {
              for (WordItem wi : wis) {
                int tag = wi.getHandle();
                double value = -Math.log(1 + wi.getFreq());
                value += Math.log(context.getFreq(0, tag) + wis.size());
                sn.addPos(new POS(tag, value));
              }
            }
          }
        }

        if (sn.getAllPos() == null)
          guessPos(tagType, sn);
       
        // ���һ���ʽڵ��Ӧ��allPosΪnull����˵�����޷������ɴ�
        // ���Ĵ�������һ���ʵĴ���
        if (i - 1 >= 0 && sns.get(i - 1).getPosSize() == -1) {
          if (sn.getPosSize() > 0) {
            POS pos = new POS(sn.getAllPos().get(0).getTag(), 0);
            sns.get(i - 1).addPos(pos);
          }
        }
      }

      // ���һ��������
      SegNode last = sns.get(i - 1);
      if (last != null) {
        SegNode sn = new SegNode();
        int tag = 0;
        if (tagType != Utility.TAG_TYPE.TT_NORMAL)
          tag = 101;
        else
          tag = 1;
        POS pos = new POS(tag, 0);
        sn.addPos(pos);
        sns.add(sn);
      }
    }
  }
View Full Code Here
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
    if (sns != null && context != null) {
      for (int i = 0; i < sns.size(); i++) {
        if (i == 0) {
          int pos = tagType != Utility.TAG_TYPE.TT_NORMAL ? 100 : 0;
          prevAllPos = new ArrayList<POS>();
          prevAllPos.add(new POS(pos, 0));
        } else {
          prevAllPos = sns.get(i - 1).getAllPos();
        }
        allPos = sns.get(i).getAllPos();
        if(allPos!=null)
        for (POS pos : allPos) {
          int minPrev = 0;
          double minFreq = 1000;
          for (int k = 0;prevAllPos!=null &&  k < prevAllPos.size(); k++) {
            POS prevPos = prevAllPos.get(k);
            double temp = context.getPossibility(0, prevPos.getTag(), pos.getTag());
            temp = -Math.log(temp) + prevPos.getFreq();
            if (temp < minFreq) {
              minFreq = temp;
              minPrev = k;
            }
          }
View Full Code Here
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
      case TT_NORMAL:
        break;
      case TT_PERSON:
        if (word.indexOf("����") != -1) {
          freq = (double) 1 / (double) (context.getFreq(0, 6) + 1);
          sn.addPos(new POS(6, freq));
        } else {
          freq = (double) 1 / (double) (context.getFreq(0, 0) + 1);
          sn.addPos(new POS(0, freq));

          if (sn.getLen() >= 4) {
            freq = (double) 1 / (double) (context.getFreq(0, 0) + 1);
            sn.addPos(new POS(0, freq));
            freq = (double) 1 / (double) (context.getFreq(0, 11) * 8);
            sn.addPos(new POS(11, freq));
            freq = (double) 1 / (double) (context.getFreq(0, 12) * 8);
            sn.addPos(new POS(12, freq));
            freq = (double) 1 / (double) (context.getFreq(0, 13) * 8);
            sn.addPos(new POS(13, freq));
          } else if (sn.getLen() == 2) {
            freq = (double) 1 / (double) (context.getFreq(0, 0) + 1);
            sn.addPos(new POS(0, freq));
            charType = Utility.charType(word);
            if (charType == Utility.CT_OTHER || charType == Utility.CT_CHINESE) {
              freq = (double) 1 / (double) (context.getFreq(0, 1) + 1);
              sn.addPos(new POS(1, freq));
              freq = (double) 1 / (double) (context.getFreq(0, 2) + 1);
              sn.addPos(new POS(2, freq));
              freq = (double) 1 / (double) (context.getFreq(0, 3) + 1);
              sn.addPos(new POS(3, freq));
              freq = (double) 1 / (double) (context.getFreq(0, 4) + 1);
              sn.addPos(new POS(4, freq));
            }
            freq = (double) 1 / (double) (context.getFreq(0, 11) * 8);
            sn.addPos(new POS(11, freq));
            freq = (double) 1 / (double) (context.getFreq(0, 12) * 8);
            sn.addPos(new POS(12, freq));
            freq = (double) 1 / (double) (context.getFreq(0, 13) * 8);
            sn.addPos(new POS(13, freq));
          }
        }
        break;
      case TT_PLACE:
        freq = (double) 1 / (double) (context.getFreq(0, 0) + 1);
        sn.addPos(new POS(0, freq));

        if (sn.getLen() >= 4) {
          freq = (double) 1 / (double) (context.getFreq(0, 11) * 8);
          sn.addPos(new POS(11, freq));
          freq = (double) 1 / (double) (context.getFreq(0, 12) * 8);
          sn.addPos(new POS(12, freq));
          freq = (double) 1 / (double) (context.getFreq(0, 13) * 8);
          sn.addPos(new POS(13, freq));
        } else if (sn.getLen() == 2) {
          freq = (double) 1 / (double) (context.getFreq(0, 0) + 1);
          sn.addPos(new POS(0, freq));
          charType = Utility.charType(word);
          if (charType == Utility.CT_OTHER || charType == Utility.CT_CHINESE) {

            freq = (double) 1 / (double) (context.getFreq(0, 1) + 1);
            sn.addPos(new POS(1, freq));
            freq = (double) 1 / (double) (context.getFreq(0, 2) + 1);
            sn.addPos(new POS(2, freq));
            freq = (double) 1 / (double) (context.getFreq(0, 3) + 1);
            sn.addPos(new POS(3, freq));
            freq = (double) 1 / (double) (context.getFreq(0, 4) + 1);
            sn.addPos(new POS(4, freq));
          }
          freq = (double) 1 / (double) (context.getFreq(0, 11) * 8);
          sn.addPos(new POS(11, freq));
          freq = (double) 1 / (double) (context.getFreq(0, 12) * 8);
          sn.addPos(new POS(12, freq));
          freq = (double) 1 / (double) (context.getFreq(0, 13) * 8);
          sn.addPos(new POS(13, freq));
        }
        break;
      case TT_TRANS_PERSON:
        freq = (double) 1 / (double) (context.getFreq(0, 0) + 1);
        sn.addPos(new POS(0, freq));
        if (!Utility.isAllChinese(word)) {
          if (Utility.isAllLetter(word)) {
            freq = (double) 1 / (double) (context.getFreq(0, 1) + 1);
            sn.addPos(new POS(1, freq));
            freq = (double) 1 / (double) (context.getFreq(0, 11) + 1);
            sn.addPos(new POS(11, freq));
            freq = (double) 1 / (double) (context.getFreq(0, 2) * 2 + 1);
            sn.addPos(new POS(2, freq));
            freq = (double) 1 / (double) (context.getFreq(0, 3) * 2 + 1);
            sn.addPos(new POS(3, freq));
            freq = (double) 1 / (double) (context.getFreq(0, 12) * 2 + 1);
            sn.addPos(new POS(12, freq));
            freq = (double) 1 / (double) (context.getFreq(0, 13) * 2 + 1);
            sn.addPos(new POS(13, freq));
          }
          freq = (double) 1 / (double) (context.getFreq(0, 41) * 8);
          sn.addPos(new POS(41, freq));
          freq = (double) 1 / (double) (context.getFreq(0, 42) * 8);
          sn.addPos(new POS(42, freq));
          freq = (double) 1 / (double) (context.getFreq(0, 43) * 8);
          sn.addPos(new POS(43, freq));
        } else if (sn.getLen() >= 4) {
          freq = (double) 1 / (double) (context.getFreq(0, 41) * 8);
          sn.addPos(new POS(41, freq));
          freq = (double) 1 / (double) (context.getFreq(0, 42) * 8);
          sn.addPos(new POS(42, freq));
          freq = (double) 1 / (double) (context.getFreq(0, 43) * 8);
          sn.addPos(new POS(43, freq));
        } else if (sn.getLen() == 2) {
          charType = Utility.charType(word);
          if (charType == Utility.CT_OTHER || charType == Utility.CT_CHINESE) {
            freq = (double) 1 / (double) (context.getFreq(0, 1) * 2 + 1);
            sn.addPos(new POS(1, freq));
            freq = (double) 1 / (double) (context.getFreq(0, 2) * 2 + 1);
            sn.addPos(new POS(2, freq));
            freq = (double) 1 / (double) (context.getFreq(0, 3) * 2 + 1);
            sn.addPos(new POS(3, freq));
            freq = (double) 1 / (double) (context.getFreq(0, 30) * 8 + 1);
            sn.addPos(new POS(30, freq));
            freq = (double) 1 / (double) (context.getFreq(0, 11) * 4 + 1);
            sn.addPos(new POS(11, freq));
            freq = (double) 1 / (double) (context.getFreq(0, 12) * 4 + 1);
            sn.addPos(new POS(12, freq));
            freq = (double) 1 / (double) (context.getFreq(0, 13) * 4 + 1);
            sn.addPos(new POS(13, freq));
            freq = (double) 1 / (double) (context.getFreq(0, 21) * 2 + 1);
            sn.addPos(new POS(21, freq));
            freq = (double) 1 / (double) (context.getFreq(0, 22) * 2 + 1);
            sn.addPos(new POS(22, freq));
            freq = (double) 1 / (double) (context.getFreq(0, 23) * 2 + 1);
            sn.addPos(new POS(23, freq));
          }
          freq = (double) 1 / (double) (context.getFreq(0, 41) * 8);
          sn.addPos(new POS(41, freq));
          freq = (double) 1 / (double) (context.getFreq(0, 42) * 8);
          sn.addPos(new POS(42, freq));
          freq = (double) 1 / (double) (context.getFreq(0, 43) * 8);
          sn.addPos(new POS(43, freq));
        }
        break;
      default:
        break;
      }
View Full Code Here
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
      // �����ǿ�ʼ�ͽ������
      for (int i = size - 1, j = 0; i >= 0; i--) {
        ArrayList<POS> allPos = sns.get(i).getAllPos();
        if (allPos != null && allPos.size() > j) {
          POS pos = allPos.get(j);
          pos.setBest(true);
          j = pos.getPrev();
        } else if (i + 1 < size - 1) {
          int tag = getBestTag(sns.get(i + 1));
          POS pos = new POS(tag, 0);
          pos.setBest(true);
          sns.get(i).addPos(pos);
        }
      }
      // �ѽ�����ȥ�����õ�����Ŀ�Ľ�����Ϊ�˵õ����һ����ĩ����ĩ���ʵ����Ŵ���
View Full Code Here
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
        sn = snList.get(i);
        posList = sn.getAllPos();
        for (int j = 0; posList != null && j < posList.size(); j++) {
          double minFee = 1000000;
          int minPrev = 100000;
          POS pos = posList.get(j);
          SegNode psn = snList.get(i - 1);
          ArrayList<POS> pposList = psn.getAllPos();
          for (int k = 0; pposList != null && k < pposList.size(); k++) {
            double temp = -Math.log(context
                .getPossibility(0, pposList.get(k).getTag(), pos.getTag()));
            temp += pposList.get(k).getFreq();// Add the fees
            if (temp < minFee) {
              minFee = temp;
              minPrev = k;
            }
          }
          pos.setPrev(minPrev);
          pos.setFreq(pos.getFreq() + minFee);
        }
      }
    }
  }
View Full Code Here
117
118
119
120
121
122
123
124
125
126
127
128
            // }
            // }
            // ֻ��һ���Դʣ�������
            SegNode sg = null;
            if (sa.getPosCount() == 1) {
              Pos pos = sa.getPos(0);
              sg = new SegNode(i, j, pos.getTag(), sa.getTotalFreq(), word);
            } else
              sg = new SegNode(i, j, 0, sa.getTotalFreq(), word);
            sg.setSrcWord(word);
            sg.setGbkID(gbkID);
            segGraph.insert(sg, true);
View Full Code Here
TOP

Related Classes of org.ictclas4j.bean.POS

  • org.ictclas4j.segment.GraphGenerate
  • org.ictclas4j.segment.PosTagger
  • org.ictclas4j.segment.SegGraph
  • org.ictclas4j.test2.DictTest
    • Copyright © 2018 www.massapicom. All rights reserved.
    All source code are property of their respective owners. Java is a trademark of Sun Microsystems, Inc and owned by ORACLE Inc. Contact coftware#gmail.com.