Multiobjective Evolutionary Algorithms applied to Feature Selection in Microarrays Cancer Data
Microarray analysis of gene expression is a current topic for the diagnosis and classification of human cancer. A gene expression data microarray consists of an array of thousands of features of which most are irrelevant for classifying patterns of gene expressions. Choosing a minimal subset of feat...
- Autores:
- Tipo de recurso:
- Article of journal
- Fecha de publicación:
- 2020
- Institución:
- Universidad Católica de Pereira
- Repositorio:
- Repositorio Institucional - RIBUC
- Idioma:
- spa
- OAI Identifier:
- oai:repositorio.ucp.edu.co:10785/13682
- Acceso en línea:
- https://revistas.ucp.edu.co/index.php/entrecienciaeingenieria/article/view/2014
http://hdl.handle.net/10785/13682
- Palabra clave:
- Rights
- openAccess
- License
- Derechos de autor 2021 Entre Ciencia e Ingeniería
| Summary: | Microarray analysis of gene expression is a current topic for the diagnosis and classification of human cancer. A gene expression data microarray consists of an array of thousands of features of which most are irrelevant for classifying patterns of gene expressions. Choosing a minimal subset of features for classification is a difficult task. In this work, a comparison is made between two multi-objective evolutionary algorithms applied to sets of gene expressions popular in the literature (lymphoma, leukemia and colon). In order to remove the strongly correlated characteristics, a pre-processing stage is performed. An extensive and detailed analysis of the results obtained for the selected multi-objective algorithms is shown. |
|---|