Abstract: Existing association rule mining algorithms suffer from
many problems when mining massive transactional datasets. Mining of associations
can be viewed as a process of discovering associations among transactions within
relational databases. Finding frequent itemsets is computationally the most
expensive step in association rule discovery and therefore it has attracted
significant research attraction. Existing frequent pattern discovery algorithms
suffer from many problems regarding the high memory depending when mining large
amount of data, computational and cost; in addition, the recursive
mining process to mine these structures is also too voracious in memory
resources. This paper presents an efficient algorithm for mining complete
frequent itemsets from typical datasets. The algorithm is partially based on
FP-tree hypothesis in which the tree is built using smaller candidate sets for
large 2-itemsets generated by using an efficient dynamic hashing technique and
extracts the possible frequent itemsets directly from the tree.
Keywords and phrases: association rules, data mining, dynamic hashing, frequent items, header table, minimum support.
