Hybridizing Cartesian Genetic Programming and Harmony Search for adaptive feature construction in supervised learning problems
- Andoni Elola,
- ,
- Miren Nekane Bilbao,
- Cristina Perfecto,
- Enrique Alexandre,
- Sancho Salcedo-Sanz
- ,
- Basque Center for Applied Mathematics,
- University of Alcalá
Publicación:
Contribución a una revista
Artículo
Investigación
Revisión por expertosAcceso abierto
Abstract
The advent of the so-called Big Data paradigm has motivated a flurry of research aimed at enhancing machine learning models by following very diverse approaches. In this context this work focuses on the automatic construction of features in supervised learning problems, which differs from the conventional selection of features in that new characteristics with enhanced predictive power are inferred from the original dataset. In particular this manuscript proposes a new iterative feature construction approach based on a self-learning meta-heuristic algorithm (Harmony Search) and a solution encoding strategy (correspondingly, Cartesian Genetic Programming) suited to represent combinations of features by means of constant-length solution vectors. The proposed feature construction algorithm, coined as Adaptive Cartesian Harmony Search (ACHS), incorporates modifications that allow exploiting the estimated predictive importance of intermediate solutions and, ultimately, attaining better convergence rate in its iterative learning procedure. The performance of the proposed ACHS scheme is assessed and compared to that rendered by the state of the art in a toy example and three practical use cases from the literature. The excellent performance figures obtained in these problems shed light on the widespread applicability of the proposed scheme to supervised learning with legacy datasets composed by already refined characteristics.
Research Output Information
Tipo de resultado
Publicación:
Contribución a una revista
Artículo
Investigación
Revisión por expertosIdioma original
InglésPáginas desde-hasta (Número de páginas)
Páginas 760-770 (11 páginas)Revista (Volumen, Número de Edición)
Applied Soft Computing (Volumen 52)Hitos de publicación
- Publicada - 2016
Estado de publicación
Publicada - 2016
ID de publicación
- Scopus: 85005915329
Publication metrics
Métricas
SciVal
FWCI
0.54
SciVal
Número de autores
6
SciVal
Percentil de artículo
55
SciVal
citaciones
16
PlumX, se abre en una pestaña nueva
Número de citas
16
Capturas
50
