Hybrid Binary Bat Algorithm with Cross-Entropy Method for Feature Selection

Feature selection aims to find an optimal subset from a given set of features. As this task can be seen as a challenging combinatorial optimization problem while the classical optimization techniques have some limitations in solving it. In this paper we propose a novel hybrid metaheuristic, improved...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:2019 4th International Conference on Control and Robotics Engineering (ICCRE) S. 165 - 169
Hauptverfasser: Li, Guocheng, Le, Chengyi
Format: Tagungsbericht
Sprache:Englisch
Veröffentlicht: IEEE 01.04.2019
Schlagworte:
bat algorithm
Classification algorithms
co-evolution
cross-entropy method
Entropy
Feature extraction
feature selection
Genetic algorithms
hybrid optimization
Optimization
Sociology
Statistics
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Feature selection aims to find an optimal subset from a given set of features. As this task can be seen as a challenging combinatorial optimization problem while the classical optimization techniques have some limitations in solving it. In this paper we propose a novel hybrid metaheuristic, improved Binary Bat Algorithm with Cross-Entropy method (BBACE), for feature selection. In the proposed BBACE algorithm, the Cross-Entropy method is embedded in Bat Algorithm to make good balance between exploitation and exploration based on co-evolution. The performance of the proposed method is evaluated on 10 standard benchmark datasets from UCI repository and compared with some well-known wrapper feature selection techniques such as GA, PSO, and ALO. The experimental results demonstrate the efficiency of the proposed approach in selecting the most informative attributes for classification and improving the classification accuracy.
DOI:10.1109/ICCRE.2019.8724270