Maximum Frequent Itemsets Discovery Algorithm Based on Granular Computing
Haicheng Chi, Zhen Liu · 2018
Different traditional frequent set discovery algorithms are applicable to data with different characteristics. Based on the summary and understanding of the current status, this paper proposes a maximum frequent itemsets discovery algorithm based on granular computing (GrGMFI). The algorithm is divided into the following stages: (1) It needs to scan the database and discover the frequent items and their corresponding bitmaps; (2)Establishing the directed hierarchy graph; (3)Mining the maximum frequent itemsets on the directed hierarchical graph. Comparing this algorithm with Mafia algorithm, we find that it has high efficiency both for dense data sets and sparse data sets.