RT Journal Article
T1 FS-Studio: An extensive and efficient feature selection experimentation tool for Weka Explorer
A1 Chacón Maldonado, Andrés Manuel
A1 Asencio Cortes, Gualberto
A1 Martínez-Álvarez, Francisco
A1 Troncoso, Alicia
K1 Feature selection
K1 Weka
K1 Parallel computation
K1 Bayesian statistical tests
AB A new tool with a friendly graphical user interface specifically designed to perform feature selection experiments in Weka Explorer allowing parallel computation is proposed in this work. The proposed tool performs Bayesian statistical tests among the selected feature selection techniques to check whether the differences are statistically significant or not. Moreover, the recently published general- purpose metaheuristic named Coronavirus Optimization Algorithm is also adapted for feature selection and integrated in the proposed tool to search for attribute subsets, allowing its use along with any Weka attribute subset evaluation algorithm. After the feature selection process is performed, both classification and regression techniques can be applied to the dataset built with the most relevant features. Finally, the output of the whole process is sent to an exportable table, customizable by means of a bar plot, in order to gather both predicted and actual values as well as the evaluation metrics.
PB Elsevier
SN 2352-7110
YR 2023
FD 2023
LK http://hdl.handle.net/10433/16557
UL http://hdl.handle.net/10433/16557
LA en
NO SoftwareX, vol 23, nº 101401, p. 1-8
NO Data Science and Big Data Lab, Universidad Pablo de Olavide
DS RIO
RD May 9, 2026