Extraction de connaissances basée sur les arbres Patricia

TitleExtraction de connaissances basée sur les arbres Patricia
Publication TypeConference Paper
Year of Publication2006
AuthorsBen Haj Hmida, M
EditorSlimani, Y
Conference NameAfrican Conference on Research in Computer Science and Applied Mathematics
Date Published11/2006
KeywordsAssociation rules, Closed Itemsets, Data mining, Frequent closed Itemsets, Frequent itemsets
Abstract

In the context of knowledge discovery, the association rules mining usually uses frequentitemsets mining technique. But the huge number of itemsets led by the mining of the complete set offrequent itemsets can be reduced to the mining of frequent closed Itemsets. The literature algorithmsproposed for this kind of items have performances that depend on the database density, dense orsparse. In this paper, we propose an algorithm for mining frequent closed itemsets based on Patriciatree structure, that permits to have good performances for the two kinds of database. Experimentalresults show that our algorithm outperforms the similar ones for most of the test databases we used.

Full Text