chaccluster.h

包含最大频繁序列的挖掘； 包含层次聚类算法

#include "CMRmisc.h"

//定义聚类的结构，包含了用户给的信息和最后的聚类结果
struct cluster_info
{
	vector<int> src_set;//input samples
	vector<int> src_set_classno; //the true class no of input sample
	int          sample_num;//number of samples
	int          feature_num;//dimention of every record,first cell of record is label!
	int			 class_num;//number of clusters
	vector<int>  des_set_classno;//output the class no of samples
	int    *eachclass_num;//number of samples in each cluster
};


class CHACCluster
{
public:
	cluster_info data;
	double **simmartix;
	int nowNum;
	vector<int> samplemapclass;
	map<int, vector<int> > classmapsample; 
	
	void compute_simailarity();
    void GetCloseted(int &clu1, int &clu2);
	void AdaptSimirityMatrix(double **p, int clu1, int clu2);
	void AdaptType(int clu1, int clu2);
	
	double cluster_time;
	CHACCluster();
	void setbit(int x, vector<int>& a);
	int bit(int x, vector<int>& a);
	void clrbit(int x, vector<int>& a );
	void ReadData(const vector<vector<int>  >& maximaloccurrenceList,
		     const vector<int>& src_set_classno,
			 const int sample_num,
			 const int feature_num,
			 const int class_num);
	double compute_sim(int sampleone, int sampletwo);
	void HACCluster();
	void GetResult(string outputFile,double feature_time);
	
};