Estudio Comparativo de Métodos de Selección de Características de Inferencia Supervisada y No Supervisada
In this work, a comparative study of feature selection methods for supervised and unsupervised inference obtained from classical PCA is presented. We deduce an expression for the cost function of PCA based on the mean square error of data and its orthonormal projection, and then this concept is exte...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Instituto Tecnológico Metropolitano
2009-12-01
|
| Series: | TecnoLógicas |
| Subjects: | |
| Online Access: | http://itmojs.itm.edu.co/index.php/tecnologicas/article/view/242 |
| _version_ | 1819027354869039104 |
|---|---|
| author | Diego H. Peluffo-Ordóñez José L. Rodríguez-Sotelo Germán Castellanos-Domínguez |
| author_facet | Diego H. Peluffo-Ordóñez José L. Rodríguez-Sotelo Germán Castellanos-Domínguez |
| author_sort | Diego H. Peluffo-Ordóñez |
| collection | DOAJ |
| description | In this work, a comparative study of feature selection methods for supervised and unsupervised inference obtained from classical PCA is presented. We deduce an expression for the cost function of PCA based on the mean square error of data and its orthonormal projection, and then this concept is extended to obtain an expression for general WPCA. Additionally, we study the supervised and unsupervised Q – α algorithm and its relation with PCA. At the end, we present results employing two data sets: A low-dimensional data set to analyze the effects of orthonormal rotation, and a highdimensional data set to assess the classification performance. The feature selection methods were assessed taking into account the number of relevant features, computational cost and classification performance. The classification was carried out using a partitional clustering algorithm. |
| first_indexed | 2024-12-21T05:41:09Z |
| format | Article |
| id | doaj.art-e69bb7db743240a497ba7bda677e0a1a |
| institution | Directory Open Access Journal |
| issn | 0123-7799 2256-5337 |
| language | English |
| last_indexed | 2024-12-21T05:41:09Z |
| publishDate | 2009-12-01 |
| publisher | Instituto Tecnológico Metropolitano |
| record_format | Article |
| series | TecnoLógicas |
| spelling | doaj.art-e69bb7db743240a497ba7bda677e0a1a2022-12-21T19:14:15ZengInstituto Tecnológico MetropolitanoTecnoLógicas0123-77992256-53372009-12-01023149166215Estudio Comparativo de Métodos de Selección de Características de Inferencia Supervisada y No SupervisadaDiego H. Peluffo-Ordóñez0José L. Rodríguez-Sotelo1Germán Castellanos-Domínguez2Ingeniera Electrónica. Estudiante de maestría en Ingeniería - Automatización Industrial. Universidad Nacional de Colombia, ManizalesIngeniero Electrónico. Estudiante de doctorado en Ingeniería – Línea de automática. Universidad Nacional de Colombia, ManizalesIngeniero en Telecomunicaciones. Ph. D. en Ingeniería. Profesor asociado al Departamento de Ingeniería Eléctrica, Electrónica y Computación de la Universidad Nacional de Colombia, MedellínIn this work, a comparative study of feature selection methods for supervised and unsupervised inference obtained from classical PCA is presented. We deduce an expression for the cost function of PCA based on the mean square error of data and its orthonormal projection, and then this concept is extended to obtain an expression for general WPCA. Additionally, we study the supervised and unsupervised Q – α algorithm and its relation with PCA. At the end, we present results employing two data sets: A low-dimensional data set to analyze the effects of orthonormal rotation, and a highdimensional data set to assess the classification performance. The feature selection methods were assessed taking into account the number of relevant features, computational cost and classification performance. The classification was carried out using a partitional clustering algorithm.http://itmojs.itm.edu.co/index.php/tecnologicas/article/view/242Proyección ortonormalPCAselección de característicasWPCA. |
| spellingShingle | Diego H. Peluffo-Ordóñez José L. Rodríguez-Sotelo Germán Castellanos-Domínguez Estudio Comparativo de Métodos de Selección de Características de Inferencia Supervisada y No Supervisada TecnoLógicas Proyección ortonormal PCA selección de características WPCA. |
| title | Estudio Comparativo de Métodos de Selección de Características de Inferencia Supervisada y No Supervisada |
| title_full | Estudio Comparativo de Métodos de Selección de Características de Inferencia Supervisada y No Supervisada |
| title_fullStr | Estudio Comparativo de Métodos de Selección de Características de Inferencia Supervisada y No Supervisada |
| title_full_unstemmed | Estudio Comparativo de Métodos de Selección de Características de Inferencia Supervisada y No Supervisada |
| title_short | Estudio Comparativo de Métodos de Selección de Características de Inferencia Supervisada y No Supervisada |
| title_sort | estudio comparativo de metodos de seleccion de caracteristicas de inferencia supervisada y no supervisada |
| topic | Proyección ortonormal PCA selección de características WPCA. |
| url | http://itmojs.itm.edu.co/index.php/tecnologicas/article/view/242 |
| work_keys_str_mv | AT diegohpeluffoordonez estudiocomparativodemetodosdeselecciondecaracteristicasdeinferenciasupervisadaynosupervisada AT joselrodriguezsotelo estudiocomparativodemetodosdeselecciondecaracteristicasdeinferenciasupervisadaynosupervisada AT germancastellanosdominguez estudiocomparativodemetodosdeselecciondecaracteristicasdeinferenciasupervisadaynosupervisada |