TY - GEN
T1 - Simultaneous features and objects selection for mixed and incomplete data
AU - Villuendas-Rey, Yenny
AU - Garcfa-Borroto, Milton
AU - Medina-Pérez, Miguel A.
AU - Ruiz-Shulcloper, José
PY - 2006
Y1 - 2006
N2 - In this paper a new simultaneous editing and feature selection method for the Most Similar Neighbor classifier is proposed. It is designed for databases with objects described by features no exclusively numeric or categorical. It is based on Testor Theory and the Compact Set Editing method, mixing edited projections until a good accuracy is achieved. Experimental results with several databases show a good performance compared to previous methods and the classifier using the original sample.
AB - In this paper a new simultaneous editing and feature selection method for the Most Similar Neighbor classifier is proposed. It is designed for databases with objects described by features no exclusively numeric or categorical. It is based on Testor Theory and the Compact Set Editing method, mixing edited projections until a good accuracy is achieved. Experimental results with several databases show a good performance compared to previous methods and the classifier using the original sample.
UR - http://www.scopus.com/inward/record.url?scp=33845228510&partnerID=8YFLogxK
U2 - 10.1007/11892755_62
DO - 10.1007/11892755_62
M3 - Contribución a la conferencia
SN - 3540465561
SN - 9783540465560
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 597
EP - 605
BT - Progress in Pattern Recognition, Image Analysis and Applications - 11th Iberoamerican Congress in Pattern Recognition, CIARP 2006, Proceedings
PB - Springer Verlag
T2 - 11th Iberoamerican Congress in Pattern Recognition, CIARP 2006
Y2 - 14 November 2006 through 17 November 2006
ER -