Testors, and particularly typical testors, have been used in feature selection and supervised classification problems. Deterministic algorithms have usually been used to find typical testors. Recently, a new approach based on evolutionary algorithms has been developed. A common problem to test the behavior of both approaches is the necessity of knowing, in advance, the number of typical testors of a given basic matrix. For an arbitrary matrix, this number can not be known unless all typical testors have been found. Therefore, this paper introduces, for the first time, a strategy to generate basic matrices for which the number of typical testors is known without to find them. This method is illustrated with some examples.
viewed = 851 times
References
A. N. Dmitriev, Y. I. Zhuravlev and F. Krendeleiev 1966. "On the mathematical principles of patterns and phenoÂmena classification." Diskretnyi Analiz. 7, 3-15.
J. Ruiz, M. Lazo, E. Alba. 2001. "An overview of the concept oftestor." Pattern Recognition. 34, 13-21.
M. Lazo, J. Ruiz. 1995. "Determining the feature releÂvance for non classically described objects and a new algorithm to compute typical fuzzy testors." Pattern ReÂcognition Letters. 16, 1259-1265.
A. Carrasco, J. Martínez 2004. "Feature selection for natural disaster texts classification Using testors." InÂtelligent Data Engineering and Automated Learning. Springer- Verlag. 8,424-429.
R. Vázquez, S. Godoy. 2007. "Using testor theory to reÂduce the dimension of neural network models." Special Issue in Neural Networks and Associative Memories. 28, 93-103.
J. Santos, A. Carrasco, J. Martínez. 2004. "Feature selecÂtion using typical testors applied to estimation of stellar parameters." Computación y Sistemas. 8, 015-023.
A. Pons, R. Gil, R. Berlanga. 2007. "Using typical tesÂtors for feature selection in text categorization." Progress in Pattern Recognition, Image Analysis and ApplicaÂtions. 12th Iberoamericann Congress on Pattern RecogÂnition. Springer Verlag.
A. Pons, J. Ruiz, R. Berlanga. 2004. "A method for the automatic summarization of topic-based clusters of doÂcuments." 8th Iberoamerican Congress on Pattern ReÂcognition. Springer Verlag. 596-603.
F. Li, Q. Zhu, X. Lin. 2009. "Topic discovery in research literature based on non-negative matrix factorization and testor theory." Asia-Pacific Conference on Information Processing. 2,266-269.
Sánchez, G. 1997. "Efficient algorithms to calculate tyÂpical testors from a basic matrix." Design and program. Master thesis. BUAP, México.
L. Morales, G. Sánchez. 2007. "FS-EX plus: A new alÂgorithm for the calculation of typical FS-testor set." ProÂgress in Pattern Recognition, Image Analysis and AppliÂcations. 12th Iberoamericann Congress on Pattern ReÂcognition. Springer Verlag.
G. Sánchez, M. Lazo, O. Fuentes. 1999. "Genetic algoÂrithm to calculate minimal typical testors." Proceedings of the IV Iberoamerican Symposium on Pattern RecogÂnition. 207-214.
E Alba, R. Santana, A. Ochoa, M. Lazo. 2000. "Finding typical testors by using an evolutionary strategy." ProÂceedings of the V Ibero American Symposium on PatÂtern Recognition. 267-278.
M. Garey, D. Johnson. 1979. "Computers and intractaÂbility: A guide to the theory of NP-completeness" W. H. Freeman and Company. New York, NY.