Towards a theory unifying implicative interestingness measures and critical values consideration in MGK
Tipo de documento
Autores
Lista de autores
Totohasina, André
Resumen
The present paper shows the possibility and the benefit to compute statistical freshold for the so-called Guillaume-Kenchaff interestingness measure MGK of association rule and compares it with other measures as Confidence, Lift and Lovinger’s one. Afterwards, it proposes a theory of normalized interestingness measure unifying a set of rule quality measures in a binary context and being surprisingly centered on MGK.
Fecha
2014
Tipo de fecha
Estado publicación
Términos clave
Enfoque
Idioma
Revisado por pares
Formato del archivo
Referencias
S. Guillaume(2000), Traitement des données volumineuses. Mesures et algorithmes d’extraction des règles d’association et règles ordinales, Universté de Nantes, Phd. X. Wu,C. Zhang,S. Zhang(2004), Efficient mining of both positive and negative association rules, ACM Transactions on information Systems, 381–405. J. Diatta, H. Ralambondrainy, A. Totohasina(2007), Towards a Unifying Probabilistic ImplicativeNormalized Quality Measure for Association Rules, Quality Measures in Data Mining book, 10, 237–250. S. Ferré(2002), Systèmes d’information logique : un paradigme logico-contextuel pour interroger, naviguer et apprendre, Université de Rennes I, Phd. R. Agrawal T. Imielinski A. Swami(1993), Mining association rules between sets of items in large databases, Proc. of the ACM SIGMOD International Conference on Management of Data, editor P. Buneman and S. Jajodia, Washington,U.S.A.. I.C. Lerman(1984),Classification et analyse ordinale des donnés, Dunod. J. Blanchard, F. Guillet,H. Briand, R. Gras(2005), Assessing rule with a probabilistic measure of deviation from equilbrium, Proc. of 11th International Symposium on Applied Stochastic Models and Data Analysis ASMDA, ENST, Brest, France, 74, 191–200. R. Gras, Almouloud S., M. Bailleul, A. Larher, M. Polo, H. Ratsimba-Rajohn, A. Totohasina(1996), L’Implication statistique. Nouvelle méthode exploratoire d’analyse des données, Coll. Recherche en Didactique des Mathématiques, Édit. La Pensée Sauvage, Grenoble. D.Sánchez, M. Vila, L. Cerdal, J.M. Serrano(2008), Association rules applied to credit card fraud detection, Expert Systems with Applications(2008), doi.10.1016/ j.eswa.2008.02.001 G. Piatetsky-Shapiro(1991), Knowledge discovery in real data bases, AI Magazine, 68–70. R. J. Hilderman,H. J. Hamilton(1999), Knowledge discovery and interestingness measures: A survey, Technical Report CS 99-04, Department of Computer Science, University of Regina. M. L. Antonie, O. R. Zaïane(2004), Mining positive and negative association rules : An approch for confined rules, Proc. 8th Int. Conf. on Principle and Practice of Knowledge Discovery in Databases (PKDD’04) 27–38. S. Brin, R. Motwani, C. Silverstein(1997), Beyond market baskets: Generalizing association rules to correlation, Proc. of the ACM SIGMOD Conference, 265–276. A. Freitas(1999) , On rule interestingness measures, Journal Knowledge-Based System, 309–315. A. Totohasina(2008), Contributions to studying association rules interestingness measures: normalization under five constraints and case study of : properties, rules composites basis and extension for applying objective in Statistic and in Physical Sciences., Habilitation to Supervizing Researchs thesis (H.D.R. degree in Malagasy system), University of Antsiranana, Madagasikara (in French) A. Totohasina, H. Ralambondrainy(2005), ION : a pertinent new measure for mining information from many types of data, Proceedings of IEEE SITIS’05, Yaoundé, Cameroon, 202–207. F. Alonso, J.P. Carença-Valente, A. L. Gonzalez, C. Montes(2002), Combining expert knowledge and data mining in medical diagnosis domain, Expert Systems with Applications 23(2002), 367-375.