EMOPG+FS: Evolutionary multi-objective prototype generation and feature selection

Issue title: Intelligent Computing for Pattern Recognition, Image Processing and Computer Vision Papers from CIARP 2014, November 2-5, 2014, Puerto Vallarta, Jalisco, Mexico

Guest editors: Eduardo Bayro-Corrochano

Article type: Research Article

Authors: Rosales-Pérez, Alejandro^{a; *} | Gonzalez, Jesus A.^b | Coello, Carlos A. Coello^c | Reyes-Garcia, Carlos A.^b | Escalante, Hugo Jair^b

Affiliations: [a] Tecnológico de Monterrey, National School in Engineering and Sciences, Monterrey, Nuevo León, Mexico | [b] Instituto Nacional de Astrofísica, Óptica y Electrónica (INAOE), Computer Science Department, Tonantzintla, Puebla, Mexico | [c] Centro de Investigación y de Estudios Avanzados del IPN (CINVESTAV-IPN), Computer Science Department, Mexico City, Mexico

Correspondence: [*] Corresponding author: Alejandro Rosales-Pérez, Tecnológico de Monterrey, Monterrey, Nuevo León, Mexico. E-mail:arosalesp@itesm.mx/arosales@inaoep.mx

Abstract: k-NN is one of the most popular and effective classifiers nowadays. However, it has some limitations that overcome its applicability in large scale scenarios: basically, it requires storing the whole training set, and it computes distances of a test sample with the training data set. These limitations have been traditionally alleviated with data reduction techniques. This paper introduces a multi-objective evolutionary approach for data reduction. Our method simultaneously generates prototypes and selects features for k-NN classifiers. Contrary to most of the existing approaches, our method treats the problem with multi-objective evolutionary optimizers. We show the effectiveness of our proposal in benchmark data and compare its performance with state of the art techniques.

Keywords: Prototype generation, feature selection, k-nearest neighbors, evolutionary multi-objective optimization

DOI: 10.3233/IDA-160844

Journal: Intelligent Data Analysis, vol. 20, no. s1, pp. S37-S51, 2016