A Tight Upper Bound on the Number of Candidate Patterns
Databases
2007-05-23 v2 Artificial Intelligence
Abstract
In the context of mining for frequent patterns using the standard levelwise algorithm, the following question arises: given the current level and the current set of frequent patterns, what is the maximal number of candidate patterns that can be generated on the next level? We answer this question by providing a tight upper bound, derived from a combinatorial result from the sixties by Kruskal and Katona. Our result is useful to reduce the number of database scans.
Cite
@article{arxiv.cs/0112007,
title = {A Tight Upper Bound on the Number of Candidate Patterns},
author = {Floris Geerts and Bart Goethals and Jan Van den Bussche},
journal= {arXiv preprint arXiv:cs/0112007},
year = {2007}
}