www.pudn.com > TextClassify.rar > CATALOGLIST.CPP
// CatalogList.cpp: implementation of the CCatalogList class.
//
//////////////////////////////////////////////////////////////////////
#include "stdafx.h"
#include "CatalogList.h"
#include "wordlist.h"
#include "direct.h"
#include "wordsegment.h"
#include "classifierparam.h"
#include "stemmer.h"
#include "classifier.h"
#include
#include
#include ".\\Utility\\Utility.h"
#ifdef _DEBUG
#define new DEBUG_NEW
#undef THIS_FILE
static char THIS_FILE[] = __FILE__;
#endif
int CCatalogList::m_nSaveMode;
char pTempStr[MAX_PATH];
char pWordTag[10];
sWeightNode *CDocNode::m_pTemp=NULL; //生成文档向量时需要使用的一块临时内存----newly added
int CDocNode::m_nAllocTempLen=0;
char CDocNode::m_pSentence[MAX_PATH*10]; //用来存放经过去掉空格回车等字母后的句子
//////////////////////////////////////////////////////////////////////
// Construction/Destruction
//////////////////////////////////////////////////////////////////////
//文档结点类
CDocNode::CDocNode(const CDocNode& x):m_nAllocLen(0),m_sWeightSet(NULL),m_pResults(NULL),m_nClassNum(0)
{
*this=x;
}
//拷贝
const CDocNode& CDocNode::operator=(const CDocNode& x)
{
if(this==&x) return *this;
m_strDocName=x.m_strDocName;
m_idxDoc=x.m_idxDoc;
if(x.m_sWeightSet!=NULL)
{
AllocBuffer(x.m_nAllocLen);
memcpy(m_sWeightSet,x.m_sWeightSet,m_nAllocLen*sizeof(sWeightNode));
}
else
{
m_sWeightSet=NULL;
m_nAllocLen=0;
}
if(x.m_pResults!=NULL)
{
AllocResultsBuffer(x.m_nClassNum);
memcpy(m_pResults,x.m_pResults,m_nClassNum*sizeof(double));
}
else
{
m_pResults=NULL;
m_nClassNum=0;
}
return *this;
}
CDocNode::CDocNode()
{
m_sWeightSet = NULL;
m_pResults = NULL;
m_nAllocLen=0;
m_nClassNum=0;
m_idxDoc=-1;
m_nCataID=-1;
}
CDocNode::~CDocNode()
{
DeallocBuffer();
DeallocResultsBuffer();
}
//nMode<=0 删除所有文档信息
//nMode>0 只删除文档向量所占用的空间
void CCatalogNode::InitCatalogNode(int nMode)
{
m_lTotalWordNum = 0;
POSITION pos_doc=m_lstDocList.GetHeadPosition();
while(pos_doc!=NULL)
{
CDocNode& docnode=m_lstDocList.GetNext(pos_doc);
docnode.DeallocBuffer();
docnode.DeallocResultsBuffer();
}
if(nMode>0) m_lstDocList.RemoveAll();
}
CCatalogNode::CCatalogNode(const CCatalogNode& x)
{
*this=x;
}
//类节点构造
const CCatalogNode& CCatalogNode::operator = (const CCatalogNode& x)
{
if(this==&x) return *this;
m_lTotalWordNum = x.m_lTotalWordNum ;
m_strCatalogName=x.m_strCatalogName;
m_strDirName=x.m_strDirName;
m_idxCata=x.m_idxCata;
m_lstDocList.RemoveAll();
POSITION pos = x.m_lstDocList.GetHeadPosition();
while(pos!=NULL)
{
CDocNode& docnode=x.m_lstDocList.GetNext(pos);
m_lstDocList.AddTail(docnode);
}
return *this;
}
const CCatalogNode& CCatalogNode::operator += (const CCatalogNode& x)
{
if(this==&x) return *this;
m_lTotalWordNum += x.m_lTotalWordNum ;
m_strCatalogName=x.m_strCatalogName;
m_strDirName=x.m_strDirName;
m_idxCata=x.m_idxCata;
POSITION pos = x.m_lstDocList.GetHeadPosition();
while(pos!=NULL)
{
CDocNode& docnode=x.m_lstDocList.GetNext(pos);
m_lstDocList.AddTail(docnode);
}
return *this;
}
CCatalogNode::CCatalogNode()
{
m_idxCata=-1;
m_lCurDocID=0;
m_lTotalWordNum=0;
InitCatalogNode();
}
CCatalogNode::~CCatalogNode()
{
InitCatalogNode();
}
void CCatalogNode::SetStartDocID(long lDocID)
{
m_lCurDocID=lDocID;
}
long CCatalogNode::ScanDirectory(CString strPath)
{
if(_chdir(strPath)) // if can't find the dir
{
CString csTmp = "目录";
csTmp+=strPath;
csTmp+="不存在!";
AfxMessageBox(csTmp);
return -1;
}
HANDLE hFinder;
LPWIN32_FIND_DATA lpFindFileData;
lpFindFileData = new WIN32_FIND_DATA;
hFinder = ::FindFirstFile("*.*",lpFindFileData );
while(::FindNextFile(hFinder,lpFindFileData))
{
if( !strcmp(lpFindFileData->cFileName,".") || !strcmp(lpFindFileData->cFileName,"..") )
continue;
if(!(lpFindFileData->dwFileAttributes & FILE_ATTRIBUTE_DIRECTORY))
{
CDocNode docnode;
docnode.m_strDocName=lpFindFileData->cFileName;
docnode.m_idxDoc=m_lCurDocID++;
docnode.m_nAllocLen=0;
docnode.m_sWeightSet=NULL;
docnode.m_nClassNum=0;
docnode.m_pResults=NULL;
AddDoc(docnode);
}
}
delete lpFindFileData;
return m_lCurDocID;
}
CCatalogList::CCatalogList()
{
m_nSaveMode=0;
}
CCatalogList::~CCatalogList()
{
}
void CCatalogList::DumpCataList(CString strFileName)
{
FILE *stream;
if( (stream = fopen( strFileName, "w+" )) == NULL )
{
AfxMessageBox("无法创建文件"+strFileName+"!");
return;
}
POSITION pos=GetFirstPosition();
while(pos!=NULL)
{
CCatalogNode& catanode=GetNext(pos);
fprintf(stream,"%d %s\n",catanode.m_idxCata,catanode.m_strCatalogName);
}
fclose(stream);
}
void CCatalogList::DumpDocList(CString strFileName)
{
FILE *stream;
if( (stream = fopen( strFileName, "w+" )) == NULL )
{
AfxMessageBox("无法创建文件"+strFileName+"!");
return;
}
POSITION pos=GetFirstPosition();
POSITION pos_doc;
while(pos!=NULL)
{
CCatalogNode& catanode=GetNext(pos);
pos_doc=catanode.GetFirstPosition();
while(pos_doc!=NULL)
{
CDocNode& docnode=catanode.GetNext(pos_doc);
if(docnode.m_nAllocLen>0)
{
fprintf(stream,"%d",catanode.m_idxCata+1);
for(int i=0;i dZero)
fprintf(stream," %d:%f",docnode.m_sWeightSet[i].s_idxWord+1,docnode.m_sWeightSet[i].s_dWeight);
}
fprintf(stream,"\n");
}
}
}
fclose(stream);
}
//nSaveMode<=0 保存文档的向量
//nSaveMode>0 不保存文档的向量
void CCatalogList::DumpToFile(CString strFileName, int nSaveMode) // view the word list content
{
CFile fBinOut;
if(!fBinOut.Open(strFileName,CFile::modeWrite | CFile::modeCreate))
{
AfxMessageBox("无法创建文件"+strFileName+"!");
return;
}
CArchive ar(&fBinOut,CArchive::store);
CCatalogList::m_nSaveMode=nSaveMode;
Serialize(ar);
ar.Close();
fBinOut.Close();
}
BOOL CCatalogList::GetFromFile(CString strFileName) // view the word list content
{
CFile fBinOut;
if(!fBinOut.Open(strFileName,CFile::modeRead))
{
AfxMessageBox("无法打开文件"+strFileName+"!");
return FALSE;
}
CArchive ar(&fBinOut,CArchive::load);
Serialize(ar);
ar.Close();
fBinOut.Close();
return TRUE;
}
//处理文档库中的文档
long CCatalogList::BuildLib(CString strDirName)
{
InitCatalogList();
return ScanDirectory(strDirName);
}
void CCatalogList::Serialize(CArchive &ar)
{
if(ar.IsStoring()) ar<>m_nSaveMode;
m_lstCatalogList.Serialize(ar);
}
void AFXAPI SerializeElements(CArchive& ar,CCatalogNode* pElements,int nCount)
{
ASSERT(nCount==0||
AfxIsValidAddress(pElements,nCount*sizeof(CCatalogNode)));
pElements->Serialize(ar);
}
void AFXAPI SerializeElements(CArchive& ar,CDocNode* pElements,int nCount)
{
ASSERT(nCount==0||
AfxIsValidAddress(pElements,nCount*sizeof(CDocNode)));
pElements->Serialize(ar);
}
void CDocNode::Serialize(CArchive &ar)
{
int nLen;
if(ar.IsStoring())
{
ar<>m_idxDoc;
ar>>m_strDocName;
if(CCatalogList::GetSaveMode()<=0)
{
ar>>nLen;
AllocBuffer(nLen);
ar.Read((void*)m_sWeightSet,m_nAllocLen*sizeof(sWeightNode));
}
else
{
m_nAllocLen=0;
m_sWeightSet=NULL;
}
m_nClassNum=0;
m_pResults=NULL;
}
}
void CCatalogNode::Serialize(CArchive &ar)
{
if(ar.IsStoring())
{
ar<>m_idxCata;
ar>>m_strDirName;
ar>>m_lTotalWordNum;
ar>>m_strCatalogName;
}
m_lstDocList.Serialize(ar);
}
//为权重分配空间
void CDocNode::AllocTempBuffer(int nLen)
{
if((nLen<=0)||(m_nAllocTempLen==nLen)) return;
if(m_pTemp!=NULL)
{
delete []m_pTemp;
m_pTemp=NULL;
}
m_pTemp=new sWeightNode[nLen];
m_nAllocTempLen=nLen;
}
//释放为权重分配的临时空间
void CDocNode::DeallocTempBuffer()
{
if(m_pTemp!=NULL)
{
delete []m_pTemp;
m_pTemp=NULL;
}
m_nAllocTempLen=0;
}
//为权重分配临时空间
void CDocNode::AllocBuffer(int nLen)
{
if((nLen<=0)||(m_nAllocLen==nLen)) return;
if(m_sWeightSet!=NULL)
{
delete []m_sWeightSet;
m_sWeightSet=NULL;
}
m_nAllocLen=nLen;
m_sWeightSet=new sWeightNode[m_nAllocLen];
}
//释放为权重分配的空间
void CDocNode::DeallocBuffer()
{
if(m_sWeightSet!=NULL)
{
delete []m_sWeightSet;
m_sWeightSet=NULL;
}
m_nAllocLen=0;
}
//为分类结果分配空间
void CDocNode::AllocResultsBuffer(short nLen)
{
if((nLen<=0)||(m_nClassNum==nLen)) return;
if(m_pResults!=NULL)
{
delete []m_pResults;
m_pResults=NULL;
}
m_nClassNum=nLen;
m_pResults=new double[m_nClassNum];
}
//释放为分类结果分配的空间
void CDocNode::DeallocResultsBuffer()
{
if(m_pResults!=NULL)
{
delete []m_pResults;
m_pResults=NULL;
}
m_nClassNum=0;
}
int CDocNode::ScanChinese(char * pPath,CWordList& wordList,int nCataNum, short idxCata)
{
CFile fin;
char *buffer;
strcpy(pTempStr,pPath);
strcat(pTempStr,"\\");
strcat(pTempStr,m_strDocName.GetBuffer(0));
if(!fin.Open(pTempStr,CFile::modeRead))
return -1;
unsigned int flen=fin.GetLength();
buffer=new char[flen+1];
flen=fin.ReadHuge(buffer,flen);
buffer[flen]='\0';
fin.Close();
int num=ScanChineseString(buffer,wordList,nCataNum,m_idxDoc,idxCata);
delete[] buffer;
return num;
}
int CDocNode::ScanChineseString(char * pPath,CWordList& wordList,int nCataNum, long idxDoc, short idxCata)
{
char *buffer=pPath;
int i,j,sum;
char *w;
//realcnt为文章中去掉停用词后剩下的总共词数
//nStart为一个句子在buffer中的开始位置
int nStart=0,nNewStart=0;
bool flag=true;
int nSentenceLen=0;
int realcnt=0;
while(buffer[nStart]!='\0')
{
flag=true;
nSentenceLen=ParseFile(buffer,nStart,nNewStart);
nStart=nNewStart;
if(nSentenceLen==0) continue;
if(m_pSentence[0]>0) //如果是一个英文单词
{
//如果英文单词的长度大于等于2,且不是数字
if((nSentenceLen>=2)&&((m_pSentence[0]<'0')||(m_pSentence[0]>'9')))
{
wordList.Add(m_pSentence,idxCata,idxDoc,nCataNum);
realcnt++;
}
}
else //如果是汉字串
{
if(nSentenceLen%2!=0) continue;
if(nSentenceLen==2) //如果是单个汉字
{
wordList.Add(m_pSentence,idxCata,idxDoc,nCataNum);
realcnt++;
}
else
{
g_wordSeg.Segment(m_pSentence);
for(i=0;i32&&c<=47)||(c>=58&&c<=64)||(c>=91&&c<=96)||(c>=123&&c<=127))
{
buffer[nFilePos]='\0';
wordLen=buffer+nFilePos-p;
if(wordLen>2)
{
if(bStem) theStemmer.stem(p,0,wordLen-1);
wordList.Add(p,idxCata,idxDoc,nCataNum);
realcnt++;
}
p=buffer+nFilePos+1;
}
nFilePos++;
}
return realcnt;
}
BOOL CDocNode::IsZero()
{
ASSERT(m_sWeightSet!=NULL);
for(int i=0;icFileName,".") || !strcmp(lpFindFileData->cFileName,"..") )
continue;
if((lpFindFileData->dwFileAttributes & FILE_ATTRIBUTE_DIRECTORY))
{
CCatalogNode catalognode;
catalognode.m_strCatalogName=lpFindFileData->cFileName;
catalognode.m_strDirName=strPath+"\\"+catalognode.m_strCatalogName;
catalognode.m_idxCata=idxCurCata++;
POSITION posTemp=AddCata(catalognode);
CCatalogNode& cataTemp=GetAt(posTemp);
cataTemp.SetStartDocID(docNum);
docNum=cataTemp.ScanDirectory(catalognode.m_strDirName);
}
}
delete lpFindFileData;
return docNum;
}
CCatalogNode CCatalogList::GetAt(POSITION pos) const
{
return m_lstCatalogList.GetAt(pos);
}
CCatalogNode& CCatalogList::GetAt(POSITION pos)
{
return m_lstCatalogList.GetAt(pos);
}
//根据标识获得类节点
CCatalogNode* CCatalogList::GetCata(short idxCata)
{
POSITION pos_cata=GetFirstPosition();
while(pos_cata!=NULL)
{
CCatalogNode& catanode=GetNext(pos_cata);
if(catanode.m_idxCata==idxCata)
return &catanode;
}
return NULL;
}
//根据文档名获得文档
CDocNode* CCatalogList::GetDocByName(CString strDocName)
{
POSITION pos_cata=GetFirstPosition();
while(pos_cata!=NULL)
{
CCatalogNode& catanode=GetNext(pos_cata);
POSITION pos_doc=catanode.GetFirstPosition();
while(pos_doc!=NULL)
{
CDocNode& docnode=catanode.GetNext(pos_doc);
if(docnode.m_strDocName==strDocName)
return &docnode;
}
}
return NULL;
}
//根据类名获得类
CCatalogNode* CCatalogList::GetCataByName(CString strCataName)
{
POSITION pos_cata=GetFirstPosition();
while(pos_cata!=NULL)
{
CCatalogNode& catanode=GetNext(pos_cata);
if(catanode.m_strCatalogName==strCataName)
return &catanode;
}
return NULL;
}
//根据类名获得类的ID
short CCatalogList::GetCataIDByName(CString strCataName)
{
POSITION pos_cata=GetFirstPosition();
while(pos_cata!=NULL)
{
CCatalogNode& catanode=GetNext(pos_cata);
if(catanode.m_strCatalogName==strCataName)
return catanode.m_idxCata;
}
return -1;
}
int CDocNode::GetWordNum()
{
int s=0;
for(int i=0;idZero) nSum++;
AllocBuffer(nSum);
nSum=0;
for(i=0;idZero)
{
m_sWeightSet[nSum].s_idxWord=m_pTemp[i].s_idxWord;
m_sWeightSet[nSum].s_tfi=m_pTemp[i].s_tfi;
m_sWeightSet[nSum].s_dWeight=m_pTemp[i].s_dWeight;
nSum++;
}
}
return nSum;
}
int CDocNode::GenDocVector(DOC &doc)
{
if(m_pTemp==NULL||m_nAllocTempLen<=0) return -1;
int i,nSum=0;
for(i=0;i0) nSum++;
}
doc.words=(SVM_WORD *)malloc(sizeof(SVM_WORD)*(nSum+12));
nSum=0;
for(i=0;i0)
{
//DOC的特征ID从1开始
(doc.words[nSum]).wnum=i+1;
(doc.words[nSum]).weight=CDocNode::m_pTemp[i].s_dWeight;
nSum++;
}
}
(doc.words[nSum]).wnum=0;
doc.docnum=-1;
return nSum;
}
int CDocNode::ScanChineseStringWithDict(char *pPath,CWordList& wordList)
{
char *buffer=pPath;
int i,j,k;
short l,n;
char gram[12];
//realcnt为文章中去掉停用词后剩下的总共词数
//nStart为一个句子在buffer中的开始位置
int realcnt=0,nStart=0,nNewStart=0;
//句子的长度
int nSentenceLen=0;
memset(m_pTemp,0,sizeof(sWeightNode)*m_nAllocTempLen);
CWordNode wordNode;
while(buffer[nStart]!='\0')
{
nSentenceLen=ParseFile(buffer,nStart,nNewStart);
nStart=nNewStart;
if(nSentenceLen==0) continue;
if(m_pSentence[0]>0) //如果是一个英文单词
{
//如果英文单词的长度大于等于2,且不是数字
if((nSentenceLen>=2)&&((m_pSentence[0]<'0')||(m_pSentence[0]>'9')))
{
if(wordList.Lookup(m_pSentence,wordNode))
{
m_pTemp[wordNode.m_nWordID].s_idxWord=wordNode.m_nWordID;
m_pTemp[wordNode.m_nWordID].s_tfi+=1;
m_pTemp[wordNode.m_nWordID].s_dWeight=wordNode.m_dWeight;
realcnt++;
}
}
}
else //如果是汉字串
{
if(nSentenceLen%2!=0) continue;
//倒着扫描句子，这样分词的准确率高一些
i=nSentenceLen;
while(i>0)
{
//最长扫描5个汉字的单词
if(i>10) k=10;
else k=i;
for(j=k;j>0;j=j-2)
{
//将Gram项拷贝到gram中
n=0;
for(l=j;l>0;l--) gram[n++]=m_pSentence[i-l];
gram[n]='\0';
if(wordList.Lookup(gram,wordNode))
{
m_pTemp[wordNode.m_nWordID].s_idxWord=wordNode.m_nWordID;
m_pTemp[wordNode.m_nWordID].s_tfi+=1;
m_pTemp[wordNode.m_nWordID].s_dWeight=wordNode.m_dWeight;
realcnt++;
i=i-j+2;
break;
}
}
i=i-2;
}
}
}
//对文档向量中的每一维进行加权
if(realcnt>0)
{
double sum=0;
for(i=0;i32&&c<=47)||(c>=58&&c<=64)||(c>=91&&c<=96)||(c>=123&&c<=127))
{
buffer[nFilePos]='\0';
wordLen=buffer+nFilePos-p;
if(wordLen>2)
{
if(bStem) theStemmer.stem(p,0,wordLen-1);
if(wordList.Lookup(p,wordNode))
{
m_pTemp[wordNode.m_nWordID].s_idxWord=wordNode.m_nWordID;
m_pTemp[wordNode.m_nWordID].s_tfi+=1;
m_pTemp[wordNode.m_nWordID].s_dWeight=wordNode.m_dWeight;
realcnt++;
}
}
p=buffer+nFilePos+1;
}
nFilePos++;
}
//对文档向量中的每一维进行加权
int i;
if(realcnt>0)
{
double sum=0;
for(i=0;i0)
{
bChinese=false;
break;
}
else
{
bChinese=true;
break;
}
}
while(pBuffer[nCurrent]!='\0')
{
bChar[0]=pBuffer[nCurrent];
if(bChar[0]>127)
{
if(!bChinese) break;
nCurrent++;
bChar[1]=pBuffer[nCurrent];
//0xA1A1为全角的空格
if((bChar[0]!=0xA1)||(bChar[1]!=0xA1))
{
//如果为"的"字,或为标点符号或其它全角字母
if(((bChar[0]==0xB5)&&(bChar[1]==0xC4))||
((bChar[0]==0xA1)&&(bChar[1]>0xA1)&&(bChar[1]<=0xFE))||
((bChar[0]==0xA2)&&(bChar[1]>=0xA1)&&(bChar[1]<=0xFC))||
((bChar[0]==0xA3)&&(bChar[1]>=0xA1)&&(bChar[1]<=0xFE))||
((bChar[0]==0xA4)&&(bChar[1]>=0xA1)&&(bChar[1]<=0xF3))||
((bChar[0]==0xA5)&&(bChar[1]>=0xA1)&&(bChar[1]<=0xF6))||
((bChar[0]==0xA6)&&(bChar[1]>=0xA1)&&(bChar[1]<=0xF5))||
((bChar[0]==0xA7)&&(bChar[1]>=0xA1)&&(bChar[1]<=0xF1))||
((bChar[0]==0xA8)&&(bChar[1]>=0x40)&&(bChar[1]<=0xE9))||
((bChar[0]==0xA9)&&(bChar[1]>=0x40)&&(bChar[1]<=0xEF))||
((bChar[0]==0xAA)&&(bChar[1]==0xA5)))
{
nCurrent++;
break;
}
//如果后半部分小于0x80，代表是一个错误的汉字
else if(bChar[1]>=0x80)
{
m_pSentence[nSum]=pBuffer[nCurrent-1];
nSum++;
m_pSentence[nSum]=pBuffer[nCurrent];
nSum++;
}
}
}
else
{
if(bChar[0]==' '||bChar[0]=='\r'||bChar[0]=='\n')
{
if(!bChinese)
{
nCurrent++;
break;
}
}
else
{
//if(bChar[0]=='!'||bChar[0]=='?'||bChar[0]==':'||
// bChar[0]==';'||bChar[0]=='.')
if((bChar[0]>32&&bChar[0]<=47)||(bChar[0]>=58&&bChar[0]<=64)||
(bChar[0]>=91&&bChar[0]<=96)||(bChar[0]>=123&&bChar[0]<=127))
{
nCurrent++;
break;
}
else if(bChinese) break;
else
{
m_pSentence[nSum]=pBuffer[nCurrent];
nSum++;
}
}
}
nCurrent++;
}
m_pSentence[nSum]='\0';
nEnd=nCurrent;
return nSum;
}
bool CCatalogList::BuildCatalogID(CCatalogList & catalogList)
{
POSITION pos=GetFirstPosition();
CString strCatalogName;
short cataID;
while(pos!=NULL)
{
CCatalogNode& cataNode=GetNext(pos);
cataID=catalogList.GetCataIDByName(cataNode.m_strCatalogName);
if(cataID<0) return false;
cataNode.m_idxCata=cataID;
}
return true;
}
bool CDocNode::IsNumber(char *p)
{
int i=0;
while(p[i]!='\0'&&p[i]>='0'&&p[i]<='9') i++;
if(p[i]=='\0') return true;
else return false;
}
//函数将smart格式的类别字符串line转换一个类别ID数组
//函数的返回值为字符串line中包含无法识别的类别总数
short CCatalogList::GetCataIDArrayFromString(char * line, CArray &aryCataID)
{
int pos=0,id,d;
short result=0;
char type[MAX_PATH];
aryCataID.RemoveAll();
while(sscanf(line+pos,"%s %d",type,&d) != EOF)
{
id=GetCataIDByName(type);
if(id>=0)
{
aryCataID.Add(id);
while(line[pos]!=';'&&line[pos]!='\r'&&line[pos]!='\0')
pos++;
if(line[pos]==';') pos=pos+2;
}
else result++;
}
return result;
}
CDocNode& CCatalogNode::GetAt(POSITION position)
{
return m_lstDocList.GetAt(position);
}
int CCatalogList::GetSaveMode()
{
return CCatalogList::m_nSaveMode;
}
double CDocNode::ComputeProbability(CWordList &wordlist,int n)
{
/*
double sum=0.0;
for(int i=0;i