An Improved Methodology for Data Mining

Mining association rules is one of the important research issues in data mining. So, many algorithms have been proposed to findassociation rules in databases. However, the efficiency of algorithms for managing large data sets in the real word needs to beimproved.one of the most popular algorithms is apriori . in this paper We show the limitation of the apriori time-wasting algorithmfor scanning the entire database and searching on the frequent itemsets , and propose an modification on apriori by reducing thatwaste time depending on checking only some records in database (by partitioning the database and using new structures namedsidevector for each cluster).Experimental results show the efficiency of the presented approach for real world transactions.

کلیدواژه ها:

data mining ، cluster table ، association rules