DocumentCode :
2745864
Title :
FOntGAR algorithm: Mining generalized association rules using fuzzy ontologies
Author :
Ayres, Rodrigo Moura Juvenil ; Santos, Marilde Terezinha Prado
Author_Institution :
Dept. of Comput. Sci., Fed. Univ. of Sao Carlos - UFSCar, Sao Carlos, Brazil
fYear :
2012
fDate :
10-15 June 2012
Firstpage :
1
Lastpage :
8
Abstract :
Most of the works in mining generalized association rules under fuzzy taxonomies are focused on the pre-processing stage, using the concept of extended transactions. A great problem of these transactions is related to the generation of huge amount of candidates. Beyond that, the inclusion of ancestors in database transactions ends up generating redundancy problems. Besides, it is possible to see that many works have directed for the question of mining fuzzy rules, exploring linguistic terms, but few approaches have explored new steps of mining process. In this sense, this paper proposes the FOntGAR (Fuzzy Ontology-based Generalized Association Rules Algorithm), a new algorithm for mining generalized association rules under all levels of fuzzy concept ontologies. In this work the generalization is made during a post-processing stage. Other relevant points of this paper are the specification of a new approach of generalization; including a new grouping rules treatment, and a new and efficient way for calculating both support and confidence of generalized rules.
Keywords :
computational linguistics; data mining; fuzzy set theory; ontologies (artificial intelligence); transaction processing; FOntGAR algorithm; data redundancy; database transaction; fuzzy concept ontology; fuzzy ontology-based generalized association rule; fuzzy rule mining; fuzzy taxonomy; grouping rule treatment; linguistic; Association rules; Dairy products; Itemsets; Ontologies; Taxonomy; Vectors; Fuzzy Ontologies; Fuzzy Taxonomies; Generalized Association Rules; Post-Processing;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Fuzzy Systems (FUZZ-IEEE), 2012 IEEE International Conference on
Conference_Location :
Brisbane, QLD
ISSN :
1098-7584
Print_ISBN :
978-1-4673-1507-4
Electronic_ISBN :
1098-7584
Type :
conf
DOI :
10.1109/FUZZ-IEEE.2012.6250804
Filename :
6250804
Link To Document :
بازگشت