Package org.apache.lucene.analysis.cn.smart.hhmm

Examples of org.apache.lucene.analysis.cn.smart.hhmm.SegToken

78
79
80
81
82
83
84
85
86
87
88
        return false; // no more sentences, end of stream!
      }
    }
   
    // There are remaining tokens from the current sentence, return the next one.
    SegToken nextWord = (SegToken) tokenIter.next();
    termAtt.setTermBuffer(nextWord.charArray, 0, nextWord.charArray.length);
    offsetAtt.setOffset(nextWord.startOffset, nextWord.endOffset);
    typeAtt.setType("word");
    return true;
  }
View Full Code Here
82
83
84
85
86
87
88
89
90
91
92
      }
    }
    // WordTokenFilter must clear attributes, as it is creating new tokens.
    clearAttributes();
    // There are remaining tokens from the current sentence, return the next one.
    SegToken nextWord = tokenIter.next();
    termAtt.copyBuffer(nextWord.charArray, 0, nextWord.charArray.length);
    if (hasIllegalOffsets) {
      offsetAtt.setOffset(tokStart, tokEnd);
    } else {
      offsetAtt.setOffset(nextWord.startOffset, nextWord.endOffset);
View Full Code Here
80
81
82
83
84
85
86
87
88
89
90
      }
    }
    // WordTokenFilter must clear attributes, as it is creating new tokens.
    clearAttributes();
    // There are remaining tokens from the current sentence, return the next one.
    SegToken nextWord = tokenIter.next();
    termAtt.setTermBuffer(nextWord.charArray, 0, nextWord.charArray.length);
    offsetAtt.setOffset(nextWord.startOffset, nextWord.endOffset);
    typeAtt.setType("word");
    return true;
  }
View Full Code Here
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
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
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
    int i = 0, j;
    int length = sentence.length();
    int foundIndex;
    int[] charTypeArray = getCharTypes(sentence);
    StringBuilder wordBuf = new StringBuilder();
    SegToken token;
    int frequency = 0; // the number of times word appears.
    boolean hasFullWidth;
    int wordType;
    char[] charArray;

    SegGraph segGraph = new SegGraph();
    while (i < length) {
      hasFullWidth = false;
      switch (charTypeArray[i]) {
        case CharType.SPACE_LIKE:
          i++;
          break;
        case CharType.HANZI:
          j = i + 1;
          wordBuf.delete(0, wordBuf.length());
          // It doesn't matter if a single Chinese character (Hanzi) can form a phrase or not,
          // it will store that single Chinese character (Hanzi) in the SegGraph.  Otherwise, it will
          // cause word division.
          wordBuf.append(sentence.charAt(i));
          charArray = new char[] { sentence.charAt(i) };
          frequency = wordDict.getFrequency(charArray);
          token = new SegToken(charArray, i, j, WordType.CHINESE_WORD,
              frequency);
          segGraph.addToken(token);

          foundIndex = wordDict.getPrefixMatch(charArray);
          while (j <= length && foundIndex != -1) {
            if (wordDict.isEqual(charArray, foundIndex) && charArray.length > 1) {
              // It is the phrase we are looking for; In other words, we have found a phrase SegToken
              // from i to j.  It is not a monosyllabic word (single word).
              frequency = wordDict.getFrequency(charArray);
              token = new SegToken(charArray, i, j, WordType.CHINESE_WORD,
                  frequency);
              segGraph.addToken(token);
            }

            while (j < length && charTypeArray[j] == CharType.SPACE_LIKE)
              j++;

            if (j < length && charTypeArray[j] == CharType.HANZI) {
              wordBuf.append(sentence.charAt(j));
              charArray = new char[wordBuf.length()];
              wordBuf.getChars(0, charArray.length, charArray, 0);
              // idArray has been found (foundWordIndex!=-1) as a prefix before. 
              // Therefore, idArray after it has been lengthened can only appear after foundWordIndex. 
              // So start searching after foundWordIndex.
              foundIndex = wordDict.getPrefixMatch(charArray, foundIndex);
              j++;
            } else {
              break;
            }
          }
          i++;
          break;
        case CharType.FULLWIDTH_LETTER:
          hasFullWidth = true;
        case CharType.LETTER:
          j = i + 1;
          while (j < length
              && (charTypeArray[j] == CharType.LETTER || charTypeArray[j] == CharType.FULLWIDTH_LETTER)) {
            if (charTypeArray[j] == CharType.FULLWIDTH_LETTER)
              hasFullWidth = true;
            j++;
          }
          // Found a Token from i to j. Type is LETTER char string.
          charArray = Utility.STRING_CHAR_ARRAY;
          frequency = wordDict.getFrequency(charArray);
          wordType = hasFullWidth ? WordType.FULLWIDTH_STRING : WordType.STRING;
          token = new SegToken(charArray, i, j, wordType, frequency);
          segGraph.addToken(token);
          i = j;
          break;
        case CharType.FULLWIDTH_DIGIT:
          hasFullWidth = true;
        case CharType.DIGIT:
          j = i + 1;
          while (j < length
              && (charTypeArray[j] == CharType.DIGIT || charTypeArray[j] == CharType.FULLWIDTH_DIGIT)) {
            if (charTypeArray[j] == CharType.FULLWIDTH_DIGIT)
              hasFullWidth = true;
            j++;
          }
          // Found a Token from i to j. Type is NUMBER char string.
          charArray = Utility.NUMBER_CHAR_ARRAY;
          frequency = wordDict.getFrequency(charArray);
          wordType = hasFullWidth ? WordType.FULLWIDTH_NUMBER : WordType.NUMBER;
          token = new SegToken(charArray, i, j, wordType, frequency);
          segGraph.addToken(token);
          i = j;
          break;
        case CharType.DELIMITER:
          j = i + 1;
          // No need to search the weight for the punctuation.  Picking the highest frequency will work.
          frequency = Utility.MAX_FREQUENCE;
          charArray = new char[] { sentence.charAt(i) };
          token = new SegToken(charArray, i, j, WordType.DELIMITER, frequency);
          segGraph.addToken(token);
          i = j;
          break;
        default:
          j = i + 1;
          // Treat the unrecognized char symbol as unknown string.
          // For example, any symbol not in GB2312 is treated as one of these.
          charArray = Utility.STRING_CHAR_ARRAY;
          frequency = wordDict.getFrequency(charArray);
          token = new SegToken(charArray, i, j, WordType.STRING, frequency);
          segGraph.addToken(token);
          i = j;
          break;
      }
    }

    // Add two more Tokens: "beginning xx beginning"
    charArray = Utility.START_CHAR_ARRAY;
    frequency = wordDict.getFrequency(charArray);
    token = new SegToken(charArray, -1, 0, WordType.SENTENCE_BEGIN, frequency);
    segGraph.addToken(token);

    // "end xx end"
    charArray = Utility.END_CHAR_ARRAY;
    frequency = wordDict.getFrequency(charArray);
    token = new SegToken(charArray, length, length + 1, WordType.SENTENCE_END,
        frequency);
    segGraph.addToken(token);

    return segGraph;
  }
View Full Code Here
66
67
68
69
70
71
72
73
74
75
76
  @Override
  protected boolean incrementWord() {
    if (tokens == null || !tokens.hasNext()) {
      return false;
    } else {
      SegToken token = tokens.next();
      clearAttributes();
      termAtt.copyBuffer(token.charArray, 0, token.charArray.length);
      offsetAtt.setOffset(correctOffset(token.startOffset), correctOffset(token.endOffset));
      typeAtt.setType("word");
      return true;
View Full Code Here
79
80
81
82
83
84
85
86
87
88
89
      }
    }
    // WordTokenFilter must clear attributes, as it is creating new tokens.
    clearAttributes();
    // There are remaining tokens from the current sentence, return the next one.
    SegToken nextWord = (SegToken) tokenIter.next();
    termAtt.setTermBuffer(nextWord.charArray, 0, nextWord.charArray.length);
    offsetAtt.setOffset(nextWord.startOffset, nextWord.endOffset);
    typeAtt.setType("word");
    return true;
  }
View Full Code Here
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
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
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
    int i = 0, j;
    int length = sentence.length();
    int foundIndex;
    int[] charTypeArray = getCharTypes(sentence);
    StringBuffer wordBuf = new StringBuffer();
    SegToken token;
    int frequency = 0; // word的出现次数
    boolean hasFullWidth;
    int wordType;
    char[] charArray;

    SegGraph segGraph = new SegGraph();
    while (i < length) {
      hasFullWidth = false;
      switch (charTypeArray[i]) {
        case CharType.SPACE_LIKE:
          i++;
          break;
        case CharType.HANZI:
          j = i + 1;
          wordBuf.delete(0, wordBuf.length());
          // 不管单个汉字能不能构成词，都将单个汉字存到segGraph中去，否则会造成分此图断字
          wordBuf.append(sentence.charAt(i));
          charArray = new char[] { sentence.charAt(i) };
          frequency = wordDict.getFrequency(charArray);
          token = new SegToken(charArray, i, j, WordType.CHINESE_WORD,
              frequency);
          segGraph.addToken(token);

          foundIndex = wordDict.getPrefixMatch(charArray);
          while (j <= length && foundIndex != -1) {
            if (wordDict.isEqual(charArray, foundIndex) && charArray.length > 1) {
              // 就是我们要找的词， 也就是说找到了从i到j的一个成词SegToken，并且不是单字词
              frequency = wordDict.getFrequency(charArray);
              token = new SegToken(charArray, i, j, WordType.CHINESE_WORD,
                  frequency);
              segGraph.addToken(token);
            }

            while (j < length && charTypeArray[j] == CharType.SPACE_LIKE)
              j++;

            if (j < length && charTypeArray[j] == CharType.HANZI) {
              wordBuf.append(sentence.charAt(j));
              charArray = new char[wordBuf.length()];
              wordBuf.getChars(0, charArray.length, charArray, 0);
              // idArray作为前缀已经找到过(foundWordIndex!=-1),
              // 因此加长过后的idArray只可能出现在foundWordIndex以后,
              // 故从foundWordIndex之后开始查找
              foundIndex = wordDict.getPrefixMatch(charArray, foundIndex);
              j++;
            } else {
              break;
            }
          }
          i++;
          break;
        case CharType.FULLWIDTH_LETTER:
          hasFullWidth = true;
        case CharType.LETTER:
          j = i + 1;
          while (j < length
              && (charTypeArray[j] == CharType.LETTER || charTypeArray[j] == CharType.FULLWIDTH_LETTER)) {
            if (charTypeArray[j] == CharType.FULLWIDTH_LETTER)
              hasFullWidth = true;
            j++;
          }
          // 找到了从i到j的一个Token，类型为LETTER的字符串
          charArray = Utility.STRING_CHAR_ARRAY;
          frequency = wordDict.getFrequency(charArray);
          wordType = hasFullWidth ? WordType.FULLWIDTH_STRING : WordType.STRING;
          token = new SegToken(charArray, i, j, wordType, frequency);
          segGraph.addToken(token);
          i = j;
          break;
        case CharType.FULLWIDTH_DIGIT:
          hasFullWidth = true;
        case CharType.DIGIT:
          j = i + 1;
          while (j < length
              && (charTypeArray[j] == CharType.DIGIT || charTypeArray[j] == CharType.FULLWIDTH_DIGIT)) {
            if (charTypeArray[j] == CharType.FULLWIDTH_DIGIT)
              hasFullWidth = true;
            j++;
          }
          // 找到了从i到j的一个Token，类型为NUMBER的字符串
          charArray = Utility.NUMBER_CHAR_ARRAY;
          frequency = wordDict.getFrequency(charArray);
          wordType = hasFullWidth ? WordType.FULLWIDTH_NUMBER : WordType.NUMBER;
          token = new SegToken(charArray, i, j, wordType, frequency);
          segGraph.addToken(token);
          i = j;
          break;
        case CharType.DELIMITER:
          j = i + 1;
          // 标点符号的weight不用查了，选个最大的频率即可
          frequency = Utility.MAX_FREQUENCE;
          charArray = new char[] { sentence.charAt(i) };
          token = new SegToken(charArray, i, j, WordType.DELIMITER, frequency);
          segGraph.addToken(token);
          i = j;
          break;
        default:
          j = i + 1;
          // 把不认识的字符当作未知串看待，例如GB2312编码之外的字符，每个字符当作一个
          charArray = Utility.STRING_CHAR_ARRAY;
          frequency = wordDict.getFrequency(charArray);
          token = new SegToken(charArray, i, j, WordType.STRING, frequency);
          segGraph.addToken(token);
          i = j;
          break;
      }
    }

    // 为segGraph增加两个新Token： "始##始","末##末"
    charArray = Utility.START_CHAR_ARRAY;
    frequency = wordDict.getFrequency(charArray);
    token = new SegToken(charArray, -1, 0, WordType.SENTENCE_BEGIN, frequency);
    segGraph.addToken(token);

    // "末##末"
    charArray = Utility.END_CHAR_ARRAY;
    frequency = wordDict.getFrequency(charArray);
    token = new SegToken(charArray, length, length + 1, WordType.SENTENCE_END,
        frequency);
    segGraph.addToken(token);

    return segGraph;
  }
View Full Code Here
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
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
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
    int i = 0, j;
    int length = sentence.length();
    int foundIndex;
    int[] charTypeArray = getCharTypes(sentence);
    StringBuilder wordBuf = new StringBuilder();
    SegToken token;
    int frequency = 0; // the number of times word appears.
    boolean hasFullWidth;
    int wordType;
    char[] charArray;

    SegGraph segGraph = new SegGraph();
    while (i < length) {
      hasFullWidth = false;
      switch (charTypeArray[i]) {
        case CharType.SPACE_LIKE:
          i++;
          break;
        case CharType.HANZI:
          j = i + 1;
          wordBuf.delete(0, wordBuf.length());
          // It doesn't matter if a single Chinese character (Hanzi) can form a phrase or not,
          // it will store that single Chinese character (Hanzi) in the SegGraph.  Otherwise, it will
          // cause word division.
          wordBuf.append(sentence.charAt(i));
          charArray = new char[] { sentence.charAt(i) };
          frequency = wordDict.getFrequency(charArray);
          token = new SegToken(charArray, i, j, WordType.CHINESE_WORD,
              frequency);
          segGraph.addToken(token);

          foundIndex = wordDict.getPrefixMatch(charArray);
          while (j <= length && foundIndex != -1) {
            if (wordDict.isEqual(charArray, foundIndex) && charArray.length > 1) {
              // It is the phrase we are looking for; In other words, we have found a phrase SegToken
              // from i to j.  It is not a monosyllabic word (single word).
              frequency = wordDict.getFrequency(charArray);
              token = new SegToken(charArray, i, j, WordType.CHINESE_WORD,
                  frequency);
              segGraph.addToken(token);
            }

            while (j < length && charTypeArray[j] == CharType.SPACE_LIKE)
              j++;

            if (j < length && charTypeArray[j] == CharType.HANZI) {
              wordBuf.append(sentence.charAt(j));
              charArray = new char[wordBuf.length()];
              wordBuf.getChars(0, charArray.length, charArray, 0);
              // idArray has been found (foundWordIndex!=-1) as a prefix before. 
              // Therefore, idArray after it has been lengthened can only appear after foundWordIndex. 
              // So start searching after foundWordIndex.
              foundIndex = wordDict.getPrefixMatch(charArray, foundIndex);
              j++;
            } else {
              break;
            }
          }
          i++;
          break;
        case CharType.FULLWIDTH_LETTER:
          hasFullWidth = true; /* intentional fallthrough */
        case CharType.LETTER:
          j = i + 1;
          while (j < length
              && (charTypeArray[j] == CharType.LETTER || charTypeArray[j] == CharType.FULLWIDTH_LETTER)) {
            if (charTypeArray[j] == CharType.FULLWIDTH_LETTER)
              hasFullWidth = true;
            j++;
          }
          // Found a Token from i to j. Type is LETTER char string.
          charArray = Utility.STRING_CHAR_ARRAY;
          frequency = wordDict.getFrequency(charArray);
          wordType = hasFullWidth ? WordType.FULLWIDTH_STRING : WordType.STRING;
          token = new SegToken(charArray, i, j, wordType, frequency);
          segGraph.addToken(token);
          i = j;
          break;
        case CharType.FULLWIDTH_DIGIT:
          hasFullWidth = true; /* intentional fallthrough */
        case CharType.DIGIT:
          j = i + 1;
          while (j < length
              && (charTypeArray[j] == CharType.DIGIT || charTypeArray[j] == CharType.FULLWIDTH_DIGIT)) {
            if (charTypeArray[j] == CharType.FULLWIDTH_DIGIT)
              hasFullWidth = true;
            j++;
          }
          // Found a Token from i to j. Type is NUMBER char string.
          charArray = Utility.NUMBER_CHAR_ARRAY;
          frequency = wordDict.getFrequency(charArray);
          wordType = hasFullWidth ? WordType.FULLWIDTH_NUMBER : WordType.NUMBER;
          token = new SegToken(charArray, i, j, wordType, frequency);
          segGraph.addToken(token);
          i = j;
          break;
        case CharType.DELIMITER:
          j = i + 1;
          // No need to search the weight for the punctuation.  Picking the highest frequency will work.
          frequency = Utility.MAX_FREQUENCE;
          charArray = new char[] { sentence.charAt(i) };
          token = new SegToken(charArray, i, j, WordType.DELIMITER, frequency);
          segGraph.addToken(token);
          i = j;
          break;
        default:
          j = i + 1;
          // Treat the unrecognized char symbol as unknown string.
          // For example, any symbol not in GB2312 is treated as one of these.
          charArray = Utility.STRING_CHAR_ARRAY;
          frequency = wordDict.getFrequency(charArray);
          token = new SegToken(charArray, i, j, WordType.STRING, frequency);
          segGraph.addToken(token);
          i = j;
          break;
      }
    }

    // Add two more Tokens: "beginning xx beginning"
    charArray = Utility.START_CHAR_ARRAY;
    frequency = wordDict.getFrequency(charArray);
    token = new SegToken(charArray, -1, 0, WordType.SENTENCE_BEGIN, frequency);
    segGraph.addToken(token);

    // "end xx end"
    charArray = Utility.END_CHAR_ARRAY;
    frequency = wordDict.getFrequency(charArray);
    token = new SegToken(charArray, length, length + 1, WordType.SENTENCE_END,
        frequency);
    segGraph.addToken(token);

    return segGraph;
  }
View Full Code Here
65
66
67
68
69
70
71
72
73
74
75
  @Override
  protected boolean incrementWord() {
    if (tokens == null || !tokens.hasNext()) {
      return false;
    } else {
      SegToken token = tokens.next();
      clearAttributes();
      termAtt.copyBuffer(token.charArray, 0, token.charArray.length);
      offsetAtt.setOffset(correctOffset(token.startOffset), correctOffset(token.endOffset));
      typeAtt.setType("word");
      return true;
View Full Code Here
84
85
86
87
88
89
90
91
92
93
94
      }
    }
    // WordTokenFilter must clear attributes, as it is creating new tokens.
    clearAttributes();
    // There are remaining tokens from the current sentence, return the next one.
    SegToken nextWord = tokenIter.next();
    termAtt.copyBuffer(nextWord.charArray, 0, nextWord.charArray.length);
    if (hasIllegalOffsets) {
      offsetAtt.setOffset(tokStart, tokEnd);
    } else {
      offsetAtt.setOffset(nextWord.startOffset, nextWord.endOffset);
View Full Code Here
TOP

Related Classes of org.apache.lucene.analysis.cn.smart.hhmm.SegToken

  • org.apache.lucene.analysis.cn.smart.hhmm.HHMMSegmenter
  • org.apache.lucene.analysis.cn.smart.HMMChineseTokenizer
  • org.apache.lucene.analysis.cn.smart.WordTokenFilter
    • 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.