Package com.stimulus.archiva.language.NGramProfile

Examples of com.stimulus.archiva.language.NGramProfile.NGramEntry

112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
            profile.load(is);
            languages.add(profile);
            supportedLanguages.add(lang);
            List ngrams = profile.getSorted();
            for (int i=0; i<ngrams.size(); i++) {
                NGramEntry entry = (NGramEntry) ngrams.get(i);
                List registered = (List) tmpIdx.get(entry);
                if (registered == null) {
                    registered = new ArrayList();
                    tmpIdx.put(entry, registered);
                }
                registered.add(entry);
                entry.setProfile(profile);
            }
            list.append(" " + lang + "(" + ngrams.size() + ")");
            is.close();
          } catch (IOException e1) {
              logger.error("failed to initialize language identifier module",e1);
          }
        }
      }
      // transform all ngrams lists to arrays for performances
      Iterator keys = tmpIdx.keySet().iterator();
      while (keys.hasNext()) {
        NGramEntry entry = (NGramEntry) keys.next();
        List l = (List) tmpIdx.get(entry);
        if (l != null) {
          NGramEntry[] array = (NGramEntry[]) l.toArray(new NGramEntry[l.size()]);
          ngramsIdx.put(entry.getSeq(), array);
        }
      }
      logger.debug(list.toString());
   
      // Create the suspect profile
View Full Code Here
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
    suspect.analyze(text);
    Iterator iter = suspect.getSorted().iterator();
    float topscore = Float.MIN_VALUE;
    String lang = null;
    HashMap scores = new HashMap();
    NGramEntry searched = null;
   
    while (iter.hasNext()) {
        searched = (NGramEntry) iter.next();
        NGramEntry[] ngrams = (NGramEntry[]) ngramsIdx.get(searched.getSeq());
        if (ngrams != null) {
            for (int j=0; j<ngrams.length; j++) {
                NGramProfile profile = ngrams[j].getProfile();
                Float pScore = (Float) scores.get(profile);
                if (pScore == null) {
                    pScore = new Float(0);
                }
                float plScore = pScore.floatValue();
                plScore += ngrams[j].getFrequency() + searched.getFrequency();
                scores.put(profile, new Float(plScore));
                if (plScore > topscore) {
                    topscore = plScore;
                    lang = profile.getName();
                }
View Full Code Here
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
            profile.load(is);
            languages.add(profile);
            supportedLanguages.add(lang);
            List ngrams = profile.getSorted();
            for (int i=0; i<ngrams.size(); i++) {
                NGramEntry entry = (NGramEntry) ngrams.get(i);
                List registered = (List) tmpIdx.get(entry);
                if (registered == null) {
                    registered = new ArrayList();
                    tmpIdx.put(entry, registered);
                }
                registered.add(entry);
                entry.setProfile(profile);
            }
            list.append(" " + lang + "(" + ngrams.size() + ")");
            is.close();
          } catch (IOException e1) {
              logger.error("failed to initialize language identifier module",e1);
          }
        }
      }
      // transform all ngrams lists to arrays for performances
      Iterator keys = tmpIdx.keySet().iterator();
      while (keys.hasNext()) {
        NGramEntry entry = (NGramEntry) keys.next();
        List l = (List) tmpIdx.get(entry);
        if (l != null) {
          NGramEntry[] array = (NGramEntry[]) l.toArray(new NGramEntry[l.size()]);
          ngramsIdx.put(entry.getSeq(), array);
        }
      }
      logger.debug(list.toString());
   
      // Create the suspect profile
View Full Code Here
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
    suspect.analyze(text);
    Iterator iter = suspect.getSorted().iterator();
    float topscore = Float.MIN_VALUE;
    String lang = null;
    HashMap scores = new HashMap();
    NGramEntry searched = null;
   
    while (iter.hasNext()) {
        searched = (NGramEntry) iter.next();
        NGramEntry[] ngrams = (NGramEntry[]) ngramsIdx.get(searched.getSeq());
        if (ngrams != null) {
            for (int j=0; j<ngrams.length; j++) {
                NGramProfile profile = ngrams[j].getProfile();
                Float pScore = (Float) scores.get(profile);
                if (pScore == null) {
                    pScore = new Float(0);
                }
                float plScore = pScore.floatValue();
                plScore += ngrams[j].getFrequency() + searched.getFrequency();
                scores.put(profile, new Float(plScore));
                if (plScore > topscore) {
                    topscore = plScore;
                    lang = profile.getName();
                }
View Full Code Here
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
            profile.load(is);
            languages.add(profile);
            supportedLanguages.add(lang);
            List ngrams = profile.getSorted();
            for (int i=0; i<ngrams.size(); i++) {
                NGramEntry entry = (NGramEntry) ngrams.get(i);
                List registered = (List) tmpIdx.get(entry);
                if (registered == null) {
                    registered = new ArrayList();
                    tmpIdx.put(entry, registered);
                }
                registered.add(entry);
                entry.setProfile(profile);
            }
            list.append(" " + lang + "(" + ngrams.size() + ")");
            is.close();
          } catch (IOException e1) {
              logger.error("failed to initialize language identifier module",e1);
          }
        }
      }
      // transform all ngrams lists to arrays for performances
      Iterator keys = tmpIdx.keySet().iterator();
      while (keys.hasNext()) {
        NGramEntry entry = (NGramEntry) keys.next();
        List l = (List) tmpIdx.get(entry);
        if (l != null) {
          NGramEntry[] array = (NGramEntry[]) l.toArray(new NGramEntry[l.size()]);
          ngramsIdx.put(entry.getSeq(), array);
        }
      }
      logger.debug(list.toString());
   
      // Create the suspect profile
View Full Code Here
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
    suspect.analyze(text);
    Iterator iter = suspect.getSorted().iterator();
    float topscore = Float.MIN_VALUE;
    String lang = null;
    HashMap scores = new HashMap();
    NGramEntry searched = null;
   
    while (iter.hasNext()) {
        searched = (NGramEntry) iter.next();
        NGramEntry[] ngrams = (NGramEntry[]) ngramsIdx.get(searched.getSeq());
        if (ngrams != null) {
            for (int j=0; j<ngrams.length; j++) {
                NGramProfile profile = ngrams[j].getProfile();
                Float pScore = (Float) scores.get(profile);
                if (pScore == null) {
                    pScore = new Float(0);
                }
                float plScore = pScore.floatValue();
                plScore += ngrams[j].getFrequency() + searched.getFrequency();
                scores.put(profile, new Float(plScore));
                if (plScore > topscore) {
                    topscore = plScore;
                    lang = profile.getName();
                }
View Full Code Here
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
            profile.load(is);
            languages.add(profile);
            supportedLanguages.add(lang);
            List ngrams = profile.getSorted();
            for (int i=0; i<ngrams.size(); i++) {
                NGramEntry entry = (NGramEntry) ngrams.get(i);
                List registered = (List) tmpIdx.get(entry);
                if (registered == null) {
                    registered = new ArrayList();
                    tmpIdx.put(entry, registered);
                }
                registered.add(entry);
                entry.setProfile(profile);
            }
            list.append(" " + lang + "(" + ngrams.size() + ")");
            is.close();
          } catch (IOException e1) {
              logger.error("failed to initialize language identifier module",e1);
          }
        }
      }
      // transform all ngrams lists to arrays for performances
      Iterator keys = tmpIdx.keySet().iterator();
      while (keys.hasNext()) {
        NGramEntry entry = (NGramEntry) keys.next();
        List l = (List) tmpIdx.get(entry);
        if (l != null) {
          NGramEntry[] array = (NGramEntry[]) l.toArray(new NGramEntry[l.size()]);
          ngramsIdx.put(entry.getSeq(), array);
        }
      }
      logger.debug(list.toString());
   
      // Create the suspect profile
View Full Code Here
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
    suspect.analyze(text);
    Iterator iter = suspect.getSorted().iterator();
    float topscore = Float.MIN_VALUE;
    String lang = null;
    HashMap scores = new HashMap();
    NGramEntry searched = null;
   
    while (iter.hasNext()) {
        searched = (NGramEntry) iter.next();
        NGramEntry[] ngrams = (NGramEntry[]) ngramsIdx.get(searched.getSeq());
        if (ngrams != null) {
            for (int j=0; j<ngrams.length; j++) {
                NGramProfile profile = ngrams[j].getProfile();
                Float pScore = (Float) scores.get(profile);
                if (pScore == null) {
                    pScore = new Float(0);
                }
                float plScore = pScore.floatValue();
                plScore += ngrams[j].getFrequency() + searched.getFrequency();
                scores.put(profile, new Float(plScore));
                if (plScore > topscore) {
                    topscore = plScore;
                    lang = profile.getName();
                }
View Full Code Here
TOP

Related Classes of com.stimulus.archiva.language.NGramProfile.NGramEntry

  • com.stimulus.archiva.language.LanguageIdentifier
    • 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.