COMPARATIVE ANALYSIS OF APRIORI AND IMPROVED APRIORI ALGORITHM

Abstract: Mining data from a large database is one of the key research area. The Improved apriori algorithm helps in mining frequent items from the large database, by applying association rules between data in order to discover Knowledge from the large database. Due to the limitation and complexity of Apriori algorithm, a lot of research is being done to formulate new algorithms which will minimize the time and the number of database scans required for Knowledge Discovery through mining Association Rule among data from a large database. This paper compares and contrasts the behavior of Improved Apriori algorithm verses the existing Apriori algorithm. The Improved Apriori algorithm reduces the number of scans required for mining the frequent items from the large database and in turn mining the association rule between data. The uppermost transaction in the resultant matrix of the Improved Apriori algorithm contains the frequent item set found out by the existing Apriori algorithm. The Improved Apriori algorithm limits to consider a transaction in which all items ar