The Community for Technology Leaders
Sixth International Conference on Data Mining (ICDM'06) (2006)
Hong Kong
Dec. 18, 2006 to Dec. 22, 2006
ISSN: 1550-4786
ISBN: 0-7695-2701-9
pp: 813-817
Vania Bogorny , Universidade Federal do Rio Grande do Sul (UFRGS), Brazil
Joao Valiati , Universidade Federal do Rio Grande do Sul (UFRGS), Brazil
Sandro Camargo , Universidade Federal do Rio Grande do Sul (UFRGS), Brazil
Paulo Engel , Universidade Federal do Rio Grande do Sul (UFRGS), Brazil
Bart Kuijpers , Hasselt University and Transnational University of Limburg, Belgium
Luis O. Alvares , Universidade Federal do Rio Grande do Sul (UFRGS), Brazil
ABSTRACT
In frequent geographic pattern mining a large amount of patterns is well known a priori. This paper presents a novel approach for mining frequent geographic patterns without associations that are previously known as non-interesting. Geographic dependences are eliminated during the frequent set generation using prior knowledge. After the dependence elimination maximal generalized frequent sets are computed to remove redundant frequent sets. Experimental results show a significant reduction of both the number of frequent sets and the computational time for mining maximal frequent geographic patterns.
INDEX TERMS
null
CITATION

S. Camargo, B. Kuijpers, P. Engel, J. Valiati, L. O. Alvares and V. Bogorny, "Mining Maximal Generalized Frequent Geographic Patterns with Knowledge Constraints," Sixth International Conference on Data Mining (ICDM'06)(ICDM), Hong Kong, 2006, pp. 813-817.
doi:10.1109/ICDM.2006.110
91 ms
(Ver 3.3 (11022016))