EvoImp: Multiple Imputation of Multi-label Classification data with a genetic algorithm
PLOS ONE · 2024
DOI: 10.1371/journal.pone.0297147 ↗Kurz gefasst
Fachartikel in der Zeitschrift PLOS ONE, 2024. Verfasst von Antonio Fernando Lavareda Jacob Junior und 4 weiteren Autor:innen. Beteiligte Einrichtung: Graduate Program in Electrical Engineering (PPGEE), Federal University of Maranhão (UFMA), São Luís, Maranhão, Brazil. Hauptthemen laut MEDLINE: Algorithms, Research Design. Die Arbeit nennt 54 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Conselho Nacional de Desenvolvimento Científico e Tecnológico.
Abstract (Originalsprache)
Missing data is a prevalent problem that requires attention, as most data analysis techniques are unable to handle it. This is particularly critical in Multi-Label Classification (MLC), where only a few studies have investigated missing data in this application domain. MLC differs from Single-Label Classification (SLC) by allowing an instance to be associated with multiple classes. Movie classification is a didactic example since it can be “drama” and “bibliography” simultaneously. One of the most usual missing data treatment methods is data imputation, which seeks plausible values to fill in the missing ones. In this scenario, we propose a novel imputation method based on a multi-objective genetic algorithm for optimizing multiple data imputations called Multiple Imputation of Multi-label Classification data with a genetic algorithm, or simply EvoImp. We applied the proposed method in multi-label learning and evaluated its performance using six synthetic databases, considering various missing values distribution scenarios. The method was compared with other state-of-the-art imputation strategies, such as K-Means Imputation (KMI) and weighted K-Nearest Neighbors Imputation (WKNNI). The results proved that the proposed method outperformed the baseline in all the scenarios by achieving the best evaluation measures considering the Exact Match, Accuracy, and Hamming Loss. The superior results were constant in different dataset domains and sizes, demonstrating the EvoImp robustness. Thus, EvoImp represents a feasible solution to missing data treatment for multi-label learning.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Autor:innen und Institutionen
- Graduate Program in Electrical Engineering (PPGEE), Federal University of Maranhão (UFMA), São Luís, Maranhão, Brazil.
- Graduate Program in Computer Engineering and Systems (PECS), State University of Maranhão (UEMA), São Luís, Maranhão, Brazil.
- Corporate ReD Headquarters Fuji Electric Co., Tokyo, Japan.
- Institute of Engineering and Geosciences, Federal University of Western Pará (UFOPA), Santarém, Pará, Brazil.
ORCID-Profile: Antonio Fernando Lavareda Jacob Junior
Förderung
- Conselho Nacional de Desenvolvimento Científico e Tecnológico
Zitieren
APA
Jacob Junior, A. F. L., do Carmo, F. A., de Santana, A. L., Santana, E. E. C., & Lobato, F. M. F. (2024). EvoImp: Multiple Imputation of Multi-label Classification data with a genetic algorithm. PLOS ONE. https://doi.org/10.1371/journal.pone.0297147
Vancouver
Jacob Junior AFL, do Carmo FA, de Santana AL, Santana EEC, Lobato FMF. EvoImp: Multiple Imputation of Multi-label Classification data with a genetic algorithm. PLOS ONE. 2024. doi:10.1371/journal.pone.0297147
BibTeX
@article{JacobJunior2024,
author = {Jacob Junior, Antonio Fernando Lavareda and do Carmo, Fabricio Almeida and de Santana, Adamo Lima and Santana, Ewaldo Eder Carvalho and Lobato, Fabio Manoel Franca},
title = {EvoImp: Multiple Imputation of Multi-label Classification data with a genetic algorithm},
journal = {PLOS ONE},
year = {2024},
doi = {10.1371/journal.pone.0297147}
}
EUVIMED