Repositorio de producción científica de la Universidad de Sevilla

Mixed integer linear programming and heuristic methods for feature selection in clustering

Opened Access Mixed integer linear programming and heuristic methods for feature selection in clustering

Citas

buscar en

Estadísticas
Icon
Exportar a
Autor: Benati, Stefano
García Quiles, Sergio
Puerto Albandoz, Justo
Departamento: Universidad de Sevilla. Departamento de Estadística e Investigación Operativa
Fecha: 2018
Publicado en: Journal of the Operational Research Society, 1-17.
Tipo de documento: Artículo
Resumen: This paper studies the problem of selecting relevant features in clustering problems, out of a data set in which many features are useless, or masking. The data set comprises a set U of units, a set V of features, a set R of (tentative) cluster centres and distances dijk for every i ∈ U, k ∈ R, j ∈ V . The feature selection problem consists of finding a subset of features Q ⊆ V such that the total sum of the distances from the units to the closest centre is minimized. This is a combinatorial optimization problem that we show to be NP-complete, and we propose two mixed integer linear programming formulations to calculate the solution. Some computational experiments show that if clusters are well separated and the relevant features are easy to detect, then both formulations can solve problems with many integer variables. Conversely, if clusters overlap and relevant features are ambiguous, then even small problems are unsolved. To overcome this difficulty, we propose two heuristic method...
[Ver más]
Tamaño: 274.4Kb
Formato: PDF

URI: https://hdl.handle.net/11441/69808

DOI: 10.1080/01605682.2017.1398206

Ver versión del editor

Mostrar el registro completo del ítem


Esta obra está bajo una Licencia Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 Internacional

Este registro aparece en las siguientes colecciones