邮件透明过滤-构造哈希字典

贝叶斯邮件过滤的第一步是进行中文分词的提取，这里涉及到的第一个技术点就是建立字典的map结构，为后面的垃圾库和和法库做好准备。这里主要介绍字典库的建立：

#include <iostream>
#include <fstream>
#include <sstream>
#include <map>
#include <stdlib.h>
using namespace std;

/*
*Reference:http://blog.csdn.net/qll125596718/article/details/8306767
*/
class CDictionary
{
public:
	CDictionary();          //The default constructor,when define the class ,call this function automatically
	~CDictionary();         //The destructor
	int FindWord(string w); //Public interface,find the word of the dictionary in the hash
private:
	string strtmp;          //Read the dictionary every line
	string word;            //save the word
	map<string, int> wordhash; 
	map<string, int >::iterator worditer; 
	typedef pair<string, int> sipair;
};

/*
*define a wordhash and map the word of wordlexicon.txt to wordhash
*/
CDictionary::CDictionary()
{
	ifstream infile("wordlexicon.txt"); // open the wordlexicon.txt
	if (!infile.is_open()) // open failure
	{
		cerr << "Unable to open input file: " << "wordlexicon"
			<< " -- bailing out!" << endl;
		exit(-1);
	}
	while (getline(infile, strtmp, '\n')) 
	{
		istringstream istr(strtmp);
		istr >> word;                    //write the word of every line 
		wordhash.insert(sipair(word, 1)); 
	}
}
CDictionary::~CDictionary()
{
}

int CDictionary::FindWord(string w)
{
	if (wordhash.find(w) != wordhash.end())
	{
		return 1;
	}
	else
	{
		return 0;
	}
}

字典的结构为

map<string, int> wordhash

这样可以将所有字典里的词语存放在wordhash里面。这里涉及到map的操作，将在下一面博客里介绍。