| Revista: | Polibits |
| Base de datos: | PERIÓDICA |
| Número de sistema: | 000402941 |
| ISSN: | 1870-9044 |
| Autores: | Sabharwal, Chaman Lal1 Anjum, Bushra2 |
| Instituciones: | 1Missouri University of Science and Technology, Rolla, Misuri. Estados Unidos de América 2Amazon Inc., San Luis Obispo, California. Estados Unidos de América |
| Año: | 2016 |
| Periodo: | Ene-Jun |
| Número: | 53 |
| Paginación: | 31-42 |
| País: | México |
| Idioma: | Inglés |
| Tipo de documento: | Artículo |
| Enfoque: | Analítico |
| Resumen en inglés | The central idea of principal component analysis (PCA) is to reduce the dimensionality of a dataset consisting of a large number of interrelated variables, while retaining as much as possible of the variation present in the dataset. In this paper, we use PCA based algorithms in two diverse genres, qualitative spatial reasoning (QSR) to achieve lossless data reduction and health informatics to achieve data reduction along with improved regression analysis respectively. In an adaptive hybrid approach, we have employed PCA to traditional regression algorithms to improve their performance and representation. This yields prediction models that have both a better fit and reduced number of attributes than those produced by using standard logistic regression alone. We present examples using both synthetic data and real health datasets from UCI Repository |
| Disciplinas: | Ciencias de la computación, Matemáticas, Bibliotecología y ciencia de la información |
| Palabras clave: | Procesamiento de datos, Matemáticas aplicadas, Tecnología de la información, Estadística, Análisis de componentes principales, Biomedicina, Informática biomédica, Reducción de datos |
| Keyword: | Computer science, Mathematics, Library and information science, Data processing, Applied mathematics, Information technology, Statistics, Principal component analysis, Biomedicine, Health informatics, Data reduction |
| Texto completo: | Texto completo (Ver PDF) |
