MAD-ARM: Mobile agent based distributed association rule mining

Rapid advancement in information technology, business applications and its data storage are distributed in nature. Due to this distributed nature of the transaction databases, distributed association rule mining plays on important role to discover the interesting association and/or correlation relationships among large set of data items. Current research on distributed association rule mining focused to improve the efficiency of the algorithm and speed up the mining process. Few researchers have focused on to improve the efficiency of association rule mining in distributed environment by deploying intelligent agents for the mining frequent itemsets and generate association rule. Existing mobile agent based distributed association rule mining framework such as IDMA, EMADS, AeMSAR suffers the communication overhead. In this paper, the proposed framework called MAD-ARM, which attempts to reduce the communication overhead.