SDFP-Growth Algorithm as a Novelty of Association Rule Mining Optimization

An essential element of association rules is the strong confidence values that depend onthe support value threshold, which determines the optimum number of datasets. The existing method fordetermining the support value threshold is carried out manually by trial and error; the user determinesa support value such as 10%, 30%, or 60% according to their instincts. If the support value thresholdis inappropriate, it produces useless frequent patterns, overburdens computer resources, and wastes time.The formula for predicting the maximum count of frequent patterns was 2n- 1, wherenis the number ofdistinct items in the dataset. This paper proposes a new SDFP-growth algorithm that does not require manualdetermination of the support threshold value. The SDFP-growth algorithm will perform dimensionalityreduction on the original dataset that will generate level 1 and level 2 smaller datasets, thus automaticallyproducing a dataset with an optimum amount of data with a minimum support value threshold. The proposedformula for predicting the maximum number of frequent patterns will become2(|A|)- 1, which is|A|willalways be smaller thann. Experiments were performed on five various datasets, which reduced the numberof data dimensions by more than 3% on the Level 1 dataset and more than 69% on the Level 2 datasetby maintaining the confidence value of the strong rules. In the execution time evaluated, we found anoptimization of more than 2% on the level 1 dataset and more than 94% on the level 2 dataset.