Abstract: Quick change in the field of information technology raises the significant issue about gigantic data storage. To build up the connection among set of data items, likewise gets to be troublesome test. To defeat these issues in large data, association rule mining does a vital part. As of late, the vast majority of the analyst's purpose of intrigue is to increase the effectiveness of association rule mining based algorithm and accordingly expanding the speed of mining. In existing framework, association rules are applied on horizontal partitioned database and such strategy is relying on upon IDMA, EMADS communication. In this paper, we consider both horizontal and vertical partitioned database, while applying association rule mining framework likewise attempting to minimize the communications overhead. Further to upgrade the security of framework, RC4 algorithm is utilized to secure the horizontal and vertical partitions of database. At last framework uses the protocol for perceiving the fake and duplicate frequent rules.

Keywords: Data mining, Apriori, Distributed databases, Frequent item set, association rules.