21-05-2014, 02:13 PM
Mining frequent patterns without candidate Generation
ABSTRACT
In this frequent pattern tree structure is presented which is an extended prefix tree structure for storing compressed, crucial information about frequent patterns and develop an efficient FP tree based mining method, FP growth for mining the complete set of frequent patterns by pattern fragment growth. Efficiency of mining is achieved with three techniques: (1) a large database is compressed into a condensed, smaller data structure, (2) the FP tree based mining adopts a pattern fragment growth method to avoid costly generation of a large number of candidate sets, and (3) a divide and conquer method id used to decompose the mining task into a set of smaller tasks for mining patterns in conditional databases, which reduces the search space. The performance study shows that the FP – growth method is efficient and scalable for mining both long and short frequent patterns