TO SECURE M-PLAYERS IN ASSOCIATION RULE MINING TECHNIQUES IN DISTRIBUTED DATABASES

In the Era, Data mining technology is one of the fastest growing technologies which are used to extract the most important relevant data from the large database. But the databases are partitioned either horizontal or vertical partitioned system, also these collections of database are divided among the several parties. Association rule mining concept is one of the most important techniques in data mining sector. Here we propose a protocol for secure association rule mining techniques in horizontally distributed database and this protocol based upon the Fast distributed Mining (FDM) algorithm which is unsecured distributed database of the Apriori Algorithm. The main ingredients of this protocol concentrates that two novel secure multi-party algorithm for mining of association rules - one computes that union of private subset hold by each interacting player and another test is inclusion of element held by one player in a subset held another. Our protocol enhanced privacy with respect to multi party algorithm. In addition, our protocol is simpler, more efficient than older algorithms in terms of communication rounds, cost and computational cost