The Studies of Mining Frequent Patterns Based on Frequent Pattern Tree

摘要

Mining frequent patterns is to discover the groups of items appearing always together excess of a user specified threshold. Many approaches have been proposed for mining frequent pattern. However, either the search space or memory space is huge, such that the performance for the previous approach degrades when the database is massive or the threshold for mining frequent patterns is low. In order to decrease the usage of memory space and speed up the mining process, we study some methods for mining frequent patterns based on frequent pattern tree. The concept of our approach is to only construct a FP-tree and traverse a subtree of the FP-tree to generate all the frequent patterns for an item without constructing any other subtrees. After traversing a subtree for an item, our approach merges and removes the subtree to reduce the FP-tree smaller and smaller. We propose four methods based on this concept and compare the four methods with the famous algorithm FP-Growth which also construct a FP-tree and recursively mines frequent patterns by building conditional FP-tree.