naivebayes.h

【机器学习】文本分类源代码 【机器学习】文本分类源代码

#ifndef __NAIVEBAYES_H_
#define __NAIVEBAYES_H_

#pragma warning( disable : 4786 )

#include "doclist.h"
#include <string>
#include <map>


using namespace std;

class CNaiveBayes
{
private:
	int				m_CatCnt;
	long			m_DocCnt;
	set<int>		setWId,setCat;
	map<int,double> mapCatId_CatProb;
	map<int,long>	mapCatId_WordCnt;
	map<int,long>	mapCatId_DocCnt;
	map<int,map<int,double > > mapCatId_WIdProb;
public:
	CNaiveBayes();
	~CNaiveBayes();
	void Train( CDocList& );
	int ClassifyDocs( DOC& test_doc );
	void ClassifyDocs( string sVectorFile, string sResultFile );
};

#endif