A New Technique to Improve the Performance of Distributed Association Rules Mining

Mining association rules in distributed environments is one of the most important problems in the field of knowledge discovery and parallel and distributed computing. Communication and computation are two important factors in distributed mining of association rules. Current proposed distributed association rules mining algorithms treat all types of frequent itemsets as being the same, while there are different types of itemsets in distributed databases, e.g., derivable and non-derivable. In this study a new technique is developed to reduce communication and computation by exploiting derivability of itemsets in distributed data. In this technique derivable frequent itemsets are mined without any communication and I/O costs. This approach can be utilized in every distributed association rules mining algorithm. Experimental evaluations on real-life datasets show the effectiveness of our technique in terms of communication and run time.

Keywords:

Distributed association rules mining , data mining , Non-derivable itemsets , Distributed deduction rules