A Parallel Distributed Galois Lattice Approach for Data Mining Based on a CORBA Infrastructure

Galois lattices are tightly connected to Formal Concept Analysis as they generate a concept hierarchy that helps structuring and clustering closed frequent item sets. When used in Data mining, such a structure can help dealing with large data sets which are very common in such a context. Constructing Galois Lattices is one of the most complex problems in FCA. Our work focuses on the generation process of the Galois Lattice which is based on a parallel distributed approach relying on a CORBA infrastructure. This contribution is part of our global framework aimed at spatial association rules discovery for geomarketing purposes destined to a telecom operator. We will adopt a generation method of association rules that exploits the same concept lattice formerly used to process the closed frequent item sets and which is inspired from the minimal generator concept. The architecture is Manager/Agent based where the agent may encapsulate different concept generation methods given that these ones respect the same concept interfaces.

[1]  Nicolas Pasquier,et al.  Closed Set Based Discovery of Small Covers for Association Rules , 1999, Proc. 15èmes Journées Bases de Données Avancées, BDA.

[2]  Charu C. Aggarwal,et al.  Proceedings of the 8th ACM SIGMOD workshop on Research issues in data mining and knowledge discovery, DMKD 2003, San Diego, California, USA, June 13, 2003 , 2003, DMKD.

[3]  Azedine Boulmakoul.  Une structure logicielle distribuée pour la découverte et la visualisation des règles d'association spatiales , 2010, EDA.

[4]  Donato Malerba,et al.  A relational perspective on spatial data mining , 2008, Int. J. Data Min. Model. Manag..

[5]  Mohammed J. Zaki,et al.  Fast vertical mining using diffsets , 2003, KDD '03.

[6]  Vicky Choi.  Faster Algorithms for Constructing a Concept (Galois) Lattice , 2006, ArXiv.

[7]  J. Bordat.  Calcul pratique du treillis de Galois d'une correspondance , 1986 .

[8]  Mohammed J. Zaki,et al.  CHARM: An Efficient Algorithm for Closed Itemset Mining , 2002, SDM.

[9]  A. Boulmakoul,et al.  Closed frequent itemsets mining and structuring association rules based on Q-analysis , 2007, 2007 IEEE International Symposium on Signal Processing and Information Technology.

[10]  Petko Valtchev,et al.  A Parallel Algorithm for Lattice Construction , 2005, ICFCA.

[11]  Mohammed J. Zaki,et al.  Efficient algorithms for mining closed itemsets and their lattice structure , 2005, IEEE Transactions on Knowledge and Data Engineering.

[12]  Tomasz Imielinski,et al.  Mining association rules between sets of items in large databases , 1993, SIGMOD Conference.

[13]  Mohammed J. Zaki,et al.  Theoretical Foundations of Association Rules , 2007 .