Meta-learning and Multi-objective Optimization to Design Ensemble of Classifiers

Neto, Antonino A. Feitosa; Canuto, Anne M. P.

doi:10.1109/bracis.2014.27

Cited by 6 publications

(11 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…There are three different ways to ensure diversity among possible initial classifiers [7]: first, using different initial parameters for individual classifiers; second, having different views of the same problem using feature selection methods; and third, using different types of individual classifiers. Moreover, manual selection of individual classifiers to combine among the selected set of initial classifiers in the ensemble system usually is very time-consuming [8].…”

Section: P Golshanrad Et Al / Predicting the Best Classifier Combination Using Meta-learning And A Genetic Algorithmmentioning

confidence: 99%

“…An ensemble system is a multi-classifier system that aims at combining different classifiers to achieve a higher level of efficiency than individual classifiers achieve [6,7,[15][16][17].…”

Section: Ensemble Systemsmentioning

confidence: 99%

“…An effective factor for the performance of an ensemble system is the diversity of the individual classifiers. As a result, using a variety of classifiers in the ensemble system can substantially improve the overall performance [3,7,16,18]. The design of an ensemble system involves two important steps:…”

Section: Ensemble Systemsmentioning

confidence: 99%

See 2 more Smart Citations

MEGA: Predicting the best classifier combination using meta-learning and a genetic algorithm

Golshanrad

Rahmani

Karimian

et al. 2021

IDA

View full text Add to dashboard Cite

Classifier combination through ensemble systems is one of the most effective approaches to improve the accuracy of classification systems. Ensemble systems are generally used to combine classifiers; However, selecting the best combination of individual classifiers is a challenging task. In this paper, we propose an efficient assembling method that employs both meta-learning and a genetic algorithm for the selection of the best classifiers. Our method is called MEGA, standing for using MEta-learning and a Genetic Algorithm for algorithm recommendation. MEGA has three main components: Training, Model Interpretation and Testing. The Training component extracts meta-features of each training dataset and uses a genetic algorithm to discover the best classifier combination. The Model Interpretation component interprets the relationships between meta-features and classifiers using a priori and multi-label decision tree algorithms. Finally, the Testing component uses a weighted k-nearest-neighbors algorithm to predict the best combination of classifiers for unseen datasets. We present extensive experimental results that demonstrate the performance of MEGA. MEGA achieves superior results in a comparison of three other methods and, most importantly, is able to find novel interpretable rules that can be used to select the best combination of classifiers for an unseen dataset.

show abstract

Section: P Golshanrad Et Al / Predicting the Best Classifier Combination Using Meta-learning And A Genetic Algorithmmentioning

confidence: 99%

“…An ensemble system is a multi-classifier system that aims at combining different classifiers to achieve a higher level of efficiency than individual classifiers achieve [6,7,[15][16][17].…”

Section: Ensemble Systemsmentioning

confidence: 99%

See 1 more Smart Citation

MEGA: Predicting the best classifier combination using meta-learning and a genetic algorithm

Golshanrad

Rahmani

Karimian

et al. 2021

IDA

View full text Add to dashboard Cite

show abstract

“…A escolha da quantidade de classificadores tem impacto na acurácia. A seleção dos classificadores bem como sua quantidade pode ser encarado como um problema de busca [14].…”

Section: Estado Da Arteunclassified

“…Para tal, deve-se ter diversidade nos classificadores do comitê [13]. Uma opção para criar estes classificadoresé a partir de algoritmos multiobjetivos, fazendo uso da noção de dominância [14].…”

Section: Introductionunclassified

Algoritmo de Seleção de Classificadores para Reconhecimento de Padrões

Silva¹,

Bastos-Filho²

2015

Anais Do 12. Congresso Brasileiro De Inteligência Computacional

View full text Add to dashboard Cite

An improved sequential recommendation model based on spatial self-attention mechanism and meta learning

Ni,

Shen,

Tang

et al. 2024

Multimed Tools Appl

View full text Add to dashboard Cite

Meta-learning and Multi-objective Optimization to Design Ensemble of Classifiers

Cited by 6 publications

References 22 publications

MEGA: Predicting the best classifier combination using meta-learning and a genetic algorithm

MEGA: Predicting the best classifier combination using meta-learning and a genetic algorithm

Algoritmo de Seleção de Classificadores para Reconhecimento de Padrões

An improved sequential recommendation model based on spatial self-attention mechanism and meta learning

Contact Info

Product

Resources

About