Synthesizing heavy association rules from different real data sources

Many large organizations have multiple databases distributed over different branches. Number of such organizations is increasing over time. Thus, it is necessary to study data mining on multiple databases. In this paper the following contributions are made: Firstly, an extended model is proposed for synthesizing global patterns from local patterns in different databases. Secondly, the notion of heavy association rule in multiple databases is introduced, and an algorithm for synthesizing such association rules in multiple databases is thus proposed. Thirdly, the notion of exceptional association rule in multiple databases is introduced, and an extension is made to the proposed algorithm to notify whether a heavy association rule is high-frequent or exceptional. We present experimental results on three real datasets. Also, we make a comparative analysis between the proposed algorithm and existing algorithm.