Resumen
Discovering All Frequent Items is one of the most important steps in the association rules mining process. Typically, the minimum support is used as a criterion for selecting an interesting itemset. There are many researchers who focus to improve the efficiency of the entire data set algorithm in various ways. For example, data reduction, structuring of new data, and search space reduction. This study analyzes the advantages and disadvantages of four-type data structure: Map Itemset - Horizontal Data, Map Itemset - Vertical Data, Map Different Set -Horizontal Data, and Map Different Set - Vertical data. The experiment was conducted with 6 datasets, which are dense and sparse datasets from the UCI standard datasets. The results show that Map Differential - Horizontal Data can reduce the size of datasets better than other techniques that use dense datasets. Map Itemset - Horizontal Data can reduce the size of datasets better than other techniques that use sparse datasets.