A New Hybrid Search Based Algorithm Using Partition-COFI Tree in Association Rule Mining

Abstract

In recent years, most of the researchers discover the association rules among itemsets for large database become popular. It is one of the techniques used to mine the database. There are several efficient algorithms are produced different search strategies for finding the frequent itemsets and those algorithms are very popular in the association rule mining. Many association rule mining algorithms suffer from many problems when mining the massive datasets. Some of the major problems are: (1) repetitive scans (2) huge computation time takes during the candidacy generation and (3) high memory space required. This paper, proposed a hybrid search algorithm are developed for mining multilevel association rules and it improve the performance of algorithm. This algorithm is named as Partition-COFI Tree i.e., PC tree. The proposed algorithm works faster compared to other algorithm. It improves the performance of search space, I/O and CPU time.