itemsets.c File Reference

Frequent/Closed itemset discovery routines. More...

#include <GKlib.h>

Classes
struct	isparams_t

Functions
void	itemsets_find_frequent_itemsets (isparams_t *params, gk_csr_t *mat, int preflen, int *prefix)
gk_csr_t *	itemsets_project_matrix (isparams_t *param, gk_csr_t *mat, int cid)
void	gk_find_frequent_itemsets (int ntrans, ssize_t *tranptr, int *tranind, int minfreq, int maxfreq, int minlen, int maxlen, void(*process_itemset)(void *stateptr, int nitems, int *itemids, int ntrans, int *transids), void *stateptr)

Detailed Description

Frequent/Closed itemset discovery routines.

This file contains the code for finding frequent/closed itemests. These routines are implemented using a call-back mechanism to deal with the discovered itemsets.

Date

6/13/2008

Author

George Karypis

Version

$Id: itemsets.c 11075 2011-11-11 22:31:52Z karypis $ 

Function Documentation

gk_find_frequent_itemsets()

void gk_find_frequent_itemsets	(	int	ntrans,
		ssize_t *	tranptr,
		int *	tranind,
		int	minfreq,
		int	maxfreq,
		int	minlen,
		int	maxlen,
		void(*)(void *stateptr, int nitems, int *itemids, int ntrans, int *transids)	process_itemset,
		void *	stateptr
	)

The entry point of the frequent itemset discovery code

itemsets_find_frequent_itemsets()

void itemsets_find_frequent_itemsets	(	isparams_t *	params,
		gk_csr_t *	mat,
		int	preflen,
		int *	prefix
	)

Prototypes for this module

The recursive routine for DFS-based frequent pattern discovery

itemsets_project_matrix()

gk_csr_t * itemsets_project_matrix	(	isparams_t *	params,
		gk_csr_t *	mat,
		int	cid
	)

This function projects a matrix w.r.t. to a particular column. It performs the following steps:

• Determines the length of each column that is remaining
• Sorts the columns in increasing length
• Creates a column-based version of the matrix with the proper column ordering and renamed rowids.