TY - JOUR

T1 - The Choice of Optimal Algorithm for Frequent Itemset Mining

AU - Busarov, Vyacheslav

AU - Grafeeva, Natalia

AU - Mikhailova, Elena

PY - 2016

Y1 - 2016

N2 - The algorithms for mining of frequent itemsets appeared in the early 1990s. This problem has an important practical application, so there have appeared a lot of new methods of finding frequent itemsets. The number of existing algorithms complicates choosing the optimal algorithm for a certain task and dataset. Twelve most widely used algorithms for mining of frequent itemsets are analyzed and compared in this article. The authors discuss the capabilities of each algorithm and the features of classes of algorithms. The results of empirical research demonstrate different behavior of classes of algorithms according to certain characteristics of datasets.

AB - The algorithms for mining of frequent itemsets appeared in the early 1990s. This problem has an important practical application, so there have appeared a lot of new methods of finding frequent itemsets. The number of existing algorithms complicates choosing the optimal algorithm for a certain task and dataset. Twelve most widely used algorithms for mining of frequent itemsets are analyzed and compared in this article. The authors discuss the capabilities of each algorithm and the features of classes of algorithms. The results of empirical research demonstrate different behavior of classes of algorithms according to certain characteristics of datasets.

KW - data·mining

KW - frequent·itemsets

KW - average cover

KW - transaction·database

U2 - 10.3233/978-1-61499-714-6-211

DO - 10.3233/978-1-61499-714-6-211

M3 - Article

VL - 291

SP - 211

EP - 224

JO - Frontiers in Artificial Intelligence and Applications

JF - Frontiers in Artificial Intelligence and Applications

SN - 0922-6389

ER -