My SciELO
Services on Demand
Article
Indicators
- Cited by SciELO
Related links
- Similars in SciELO
Share
Revista Cubana de Ciencias Informáticas
On-line version ISSN 2227-1899
Abstract
R. GONZALEZ, Héctor; MORELL, Carlos and BLANCO, Antonio. Local linear regression with reduce rank for multitarget regression. Rev cuba cienc informat [online]. 2016, vol.10, n.4, pp. 184-193. ISSN 2227-1899.
ABSTRACT The purposes of this work is to study Local Weighted Regression LWR algorithm for multi-target prediction problems. The idea to estimate, the parameters of multivariate linear regression trough to singular value decomposition and reduced rank show stable solution to drive the predicting performance of this algorithm. The experimental results show that LWR is a competitive algorithm, in context to instance based learning, for multi-target prediction problems. The preliminary results are the started point for futures adaptations, to this algorithms, take into account the interdependency between outputs variables.
Keywords : KNN; LWR; Linear Regression; Multi-target Regression; Multi-Output Learning.