ALOA2i : OPTIMISATION D'EXTRACTION DES K- ITEMSETS FREQUENTS (POUR K ≤ 2)

Abstract : .In this article, we propose a novel optimization approach to the reference algorithm APRIORI (AGR 94).The approach used is based on sets one and two items. We start by calculating the supports of 1-itemsets (sets of singletons), then we prune the infrequent 1-itemsets and only keep those that are common (that is to say those with frequencies of occurrence called media whose values are greater than or equal to a minimum threshold). During the second iteration, we sort the frequent 1-itemsets in descending order of their respective holders and then we train 2-itemsets. This way association rules are discovered more quickly. Experimentally, the comparison of our algorithm with APRIORI, PASCAL, CLOSE and MAX-MINER, shows effectiveness to weakly correlated data.
Document type :
Preprints, Working Papers, ...
Complete list of metadatas

https://hal-auf.archives-ouvertes.fr/hal-01423822
Contributor : Claude Issa Nombré <>
Submitted on : Saturday, December 31, 2016 - 6:09:39 PM
Last modification on : Thursday, January 5, 2017 - 9:22:21 AM
Long-term archiving on : Saturday, April 1, 2017 - 12:12:23 PM

Files

ARTICLE_NCI.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-01423822, version 1

Collections

Citation

Claude Issa Nombré, Konan Marcelin Brou, Kouadio Prosper Kimou. ALOA2i : OPTIMISATION D'EXTRACTION DES K- ITEMSETS FREQUENTS (POUR K ≤ 2) . 2016. ⟨hal-01423822⟩

Share

Metrics

Record views

105

Files downloads

258