Package org.fnlp.ml.types

Examples of org.fnlp.ml.types.Dictionary

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
    System.out.println(tag.getSupportedTags().size());
    s = tag.tag(str);
    System.out.println(s);   
    System.out.println();
   
    CWSTagger cws2 = new CWSTagger("../models/seg.m", new Dictionary("../models/dict.txt"));
   
    //bool值指定该dict是否用于cws分词（分词和词性可以使用不同的词典）
    tag = new POSTagger(cws2, "../models/pos.m"
        , new Dictionary("../models/dict.txt"), true);//true就替换了之前的dict.txt
    tag.removeDictionary(false);//不移除分词的词典
    tag.setDictionary(new Dictionary("../models/dict.txt"), false);//设置POS词典，分词使用原来设置
   
    String str2 = "媒体计算研究所成立了，高级数据挖掘很难。乐phone很好！";
    String s2 = tag.tag(str2);
    System.out.println("处理未分词的句子，使用词典");
    System.out.println(s2);
    System.out.println();
   
    Dictionary dict = new Dictionary();
    dict.add("媒体计算","mypos1","mypos2");
    dict.add("乐phone","专有名");
    tag.setDictionary(dict, true);
    String s22 = tag.tag(str2);
    System.out.println(s22);
    System.out.println();
   
    POSTagger tag1 = new POSTagger("../models/pos.m");
    String str1 = "媒体计算 研究所 成立 了 , 高级 数据挖掘 很 难";
    String[] w = str1.split(" ");
    String[] s1 = tag1.tagSeged(w);
    System.out.println("直接处理分好词的句子:++++++++++");
    for(int i=0;i<s1.length;i++){
      System.out.print(w[i]+"/"+s1[i]+" ");
    }
    System.out.println("\n");
   
    POSTagger tag3 = new POSTagger("../models/pos.m", new Dictionary("../models/dict.txt"));
    String str3 = "媒体计算 研究所 成立 了 , 高级 数据挖掘 很 难 ";
    String[] w3 = str3.split(" ");
    String[] s3 = tag3.tagSeged(w3);
    System.out.println("直接处理分好词的句子，使用词典");
    for(int i=0;i<s3.length;i++){
View Full Code Here
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
    System.out.println("\n设置临时词典：");
    ArrayList<String> al = new ArrayList<String>();
    al.add("数据挖掘");
    al.add("媒体计算研究所");
    al.add("乐phone");
    Dictionary dict = new Dictionary(false);
    dict.addSegDict(al);
    tag.setDictionary(dict);
    s = tag.tag(str);
    System.out.println(s);
   
   
    CWSTagger tag2 = new CWSTagger("../models/seg.m", new Dictionary("../models/dict.txt"));
    System.out.println("\n使用词典的分词：");
    String str2 = "媒体计算研究所成立了, 高级数据挖掘很难。 乐phone热卖！";
    String s2 = tag2.tag(str2);
    System.out.println(s2);
   
    //使用不严格的词典
    CWSTagger tag3 = new CWSTagger("../models/seg.m", new Dictionary("../models/dict_ambiguity.txt",true));
    //尽量满足词典，比如词典中有“成立”“成立了”和“了”, 会使用Viterbi决定更合理的输出
    System.out.println("\n使用不严格的词典的分词：");
    String str3 = "媒体计算研究所成立了, 高级数据挖掘很难";
    String s3 = tag3.tag(str3);
    System.out.println(s3);
View Full Code Here
132
133
134
135
136
137
138
139
140
141
  public void setDictionary(THashSet<String> newset) {
    if(newset.size()==0)
      return;
    ArrayList<String> al = new ArrayList<String>();
    MyCollection.TSet2List(newset, al);
    Dictionary dict = new Dictionary();
    dict.addSegDict(al);
    setDictionary(dict);

  }
View Full Code Here
36
37
38
39
40
41
42
43
44
45
46
47
   * @throws IOException
   * @throws 
   */
  public static void main(String[] args) throws Exception {
    CWSTagger tag = new CWSTagger("./models/seg.m");
    Dictionary dict=new Dictionary();
    dict.addFile("./models/dict.txt");
    tag.setDictionary(dict);
    ArrayList<String> str = MyCollection.loadList("./testcase/test case seg.txt",null);
    for(String s:str){     
      String t = tag.tag(s);
//      t = tag.tag(t);
View Full Code Here
48
49
50
51
52
53
54
55
56
57
58
   
    ArrayList<String> al = new ArrayList<String>();
    al.add("兴化市");
    al.add("沈伦镇");
    al.add("樊荣村委会");
    Dictionary dict = new Dictionary(false);
    dict.addSegDict(al);
    tag.setDictionary(dict);
    s = tag.tag(str);
    assertTrue(s.equals("江苏省 兴化市 沈伦镇 樊荣村委会"));
  }
View Full Code Here
TOP

Related Classes of org.fnlp.ml.types.Dictionary

  • org.fnlp.demo.nlp.ChineseWordSegmentation
  • org.fnlp.demo.nlp.PartsOfSpeechTag
  • org.fnlp.nlp.cn.tag.CWSTagger
  • org.fnlp.nlp.cn.tag.CWSTaggerTest
  • org.fnlp.nlp.tag.TestDictSEG
  • java.io.FileInputStream
  • java.util.Scanner
  • org.fnlp.util.exception.LoadModelException
    • 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.