Author(s): Sadeq Darrab, Belgin Ergenç
Abstract: Traditional methods use a single minimum support threshold to find out the complete set of frequent patterns. However, in real word applications, using single minimum item support threshold is not adequate since it does not reflect the nature of each item. If single minimum support threshold is set too low, a huge amount of patterns are generated including uninteresting patterns. On the other hand if it is set too high, many of interesting patterns (called rare items) may be lost. Recently, several methods have been studied to tackle the rare item problem by avoiding using single minimum item support threshold. The nature of each item is considered where different items are specified with different Minimum Item Support thresholds (MIS) instead of using single support threshold. By this, the complete set of frequent patterns is generated without creating uninteresting patterns and losing substantial patterns. In this paper, we propose an efficient method, Multiple Item Support Frequent Pattern growth algorithm, MISFP-growth, to mine the complete set of frequent patterns with multiple item support thresholds. In this method, Multiple Item Support Frequent Pattern Tree, MISFP-Tree, is constructed to store all crucial information to mine frequent patterns. Since the construction of the MISFP-Tree is done with respect to minimum of MIS; post pruning and reconstruction phases are not required. In order to show the efficiency of the proposed method, it is compared with a recent tree-based algorithm, CFP-growth++ and various experiments are conducted on both real and synthetic datasets. Experimental results reveal that MISFP-growth outperforms in terms of execution time and memory space while we vary MIS values of items.
Keywords: Association rule mining, Frequent patterns, Rare itemsets, Multiple support thresholds
Pages: 1-10WSEAS Transactions on Computer Research, ISSN / E-ISSN: 1991-8755 / 2415-1521, Volume 4, 2016, Art. #1