fmmsegment.java~14~

基于朴素贝叶斯算法实现的中文文本分类程序。可以对中文文本进行分类识别

import java.lang.*;
import java.io.*;
import java.util.*;

public class FMMSegment
{
        Dictionary dic;
        int totleNumber;   //记录文中总共词汇数
        HashMap vocabulary;   //记录从文本中获取的中文词

        public FMMSegment()
        {
        }

        public FMMSegment(Dictionary newDic)
        {
                dic = newDic;
                totleNumber=0;
                vocabulary=new HashMap();
        }

        public int wordSegment(String Sentence)//中文分词
        {
                int senLen = Sentence.length();
                int i=0, j=0;
                int M=12;
                String word;
                boolean bFind = false;
                BufferedWriter out;

                while(i < senLen)
                {
                        int N= i+M<senLen ? i+M : senLen+1;
                        bFind=false;
                        for(j=N-1; j>i; j--)
                        {
                                word = Sentence.substring(i, j);
                                if(dic.Find(word))
                                {
                                  if(j>i+1)
                                  {
                                    vocabulary.put(word,new Integer(0));
                                    totleNumber++;    //累加总词汇数
                                    //将获取的单词写入文件
                                    try
                                    {
                                      out=new  BufferedWriter(new FileWriter("vocabulary.txt"));
                                      out.write(word);
                                      out.newLine();

                                    }
                                    catch (IOException e)
                                    {
                                      System.out.println(e);
                                    }

                                    System.out.print(word + " ");
                                  }
                                        bFind=true;
                                        i=j;
                                        break;
                                }
                        }
                        if(bFind == false)
                        {
                                i=j+1;
                        }
                }
                System.out.println();
                out.flush();
                return 1;
        }

        public void fileSegment(String fileName)//按行读入
        {
                try
                {
                        BufferedReader in = new BufferedReader(
                                new FileReader(fileName) );
                        String s;
                        while((s = in.readLine()) != null)
                        {
                                wordSegment(s);
                        }
                }
                catch (IOException e)
                {
                        System.out.println(e);
                }
        }

        public int NumOfVoc(){
          return totleNumber;
        }
}