BIBLIOTECA INFOTEC Centro de Investigación e Innovación en Tecnologías de la Información y Comunicación

Normal view MARC view ISBD view

A Family of Classifiers based on Feature Space Transformations and Model Selection

by Ortiz Bejar, José [Sustentante]; Téllez Avila, Eric S [Asesor]; Graff Guerrero, Mario [Co-Asesor].
Material type: materialTypeLabelBookPublisher: Aguascalientes, 2020Description: 135 páginas, 22 x 17 cm.Subject(s): Funciones de Kernel | Ciencia de Datos | Tecnologías de la información y comunicaciónOnline resources: Proyecto de titulación disponible en PDF en el Repositorio Institucional
Contents:
Improving the performance of classifiers is the realm of feature mapping, prototype selection, and kernel function transformations; these techniques aim for reducing the complexity, and also, improving the accuracy of models. In particular, the research’s objective is to combine them to transform data’s shape into another more convenient distribution; such that some simple algorithms, such as Naïve Bayes and k-Nearest Neighbors, can produce competitive classifiers. In this work, we introduce a family of classifiers based on feature mapping and kernel functions, orchestrated by simple a model selection scheme that achieves excel in performance. We provide an extensive experimental comparison of our methods with sixteen popular classifiers over different datasets supporting our claims. In addition to their competitive performance, our statistical tests also found that our methods are statistically different among them, and thus, an effective family of classifiers.
Dissertation note: Tesis que para obtener el grado de Doctor en Ciencias en Ciencia de Datos
Tags from this library: Add tag(s)
Log in to add tags.
    average rating: 0.0 (0 votes)
Item type Location Collection Call number Status Date due Barcode
Tesis Tesis Sección Tesis Trabajos de Titulación INFOTEC INFDCCD00001 O78 2020 (Browse shelf) No para préstamo (Consulta únicamente en sala) AGS20070096
Tesis Tesis Sección Tesis Trabajos de Titulación INFOTEC INFDCCD00001 O78 2020 (Browse shelf) No para préstamo (Consulta únicamente en sala) DF-TLALPAN20070095

Tesis que para obtener el grado de Doctor en Ciencias en Ciencia de Datos

Improving the performance of classifiers is the realm of feature mapping, prototype selection, and kernel function transformations; these techniques aim for reducing the complexity, and also, improving the accuracy of models. In particular, the research’s objective is to combine them to transform data’s shape into another more convenient distribution; such that some simple algorithms, such as Naïve Bayes and k-Nearest Neighbors, can produce competitive classifiers. In this work, we introduce a family of classifiers based on feature mapping and kernel functions, orchestrated by simple a model selection scheme that achieves excel in performance. We provide an extensive experimental comparison of our methods with sixteen popular classifiers over different datasets supporting our claims. In addition to their competitive performance, our
statistical tests also found that our methods are statistically different among them, and thus, an effective family of classifiers.

Click on an image to view it in the image viewer

INFOTEC Av. San Fernando 37, Col. Toriello Guerra, Tlalpan, C.P. 14050, México D.F. Copyright © 2019 Derechos reservados Tel: +52 (55) 56242800