Proposed algorithm for frequent item set generation

Data mining is an efficient technology to discover patterns in large databases. Association rule mining techniques are used to find the correlation between the various item sets in the database, and this correlation between various item sets are used in decision making and pattern analysis. In recent years the problem of finding frequent items and association rules from large datasets has been proposed by many researchers. Various research papers on association rule mining (ARM) are studied and analyzed first to understand the existing algorithms. The Apriori algorithm is the basic ARM algorithm, but it requires so many database scans to find frequent items. In Dynamic Item set counting (DIC) algorithm less number of database scans are needed, but complex data structure lattice is used. The main focus of this paper is to propose a new optimized algorithm (FI-generator) and to compare its performance with the existing algorithms. A secondary data set is used to find out frequent item sets and association rules with the help of existing and proposed algorithm). We observed that the proposed algorithm find out the frequent item sets and association rules from databases as compared to the existing algorithms in less numbers of database scans. In the proposed algorithm an optimized data structure adjacency matrix is used. Proposed algorithm reduces the size of candidate-K item set in successive iteration. Pruning is also done at two stages which reduces the memory space.