A novel genetic algorithm approach for simultaneous feature and classifier selection in multi classifier system

Nguyen, Tien Thanh; Liew, Alan Wee‐Chung; Tran, Minh Toan; Pham, Xuan Cuong; Nguyen, Mai P.

doi:10.1109/cec.2014.6900377

Cited by 30 publications

(20 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A poor selection of learning algorithms may result in the poor performance of the ensemble. The proposed method could be combined with learning algorithm selection [25] to acquire the optimal set of learning algorithms for each specific dataset.…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Combining heterogeneous classifiers via granular prototypes

Nguyen

Pham

et al. 2018

Applied Soft Computing

Self Cite

View full text Add to dashboard Cite

In this study, a novel framework to combine multiple classifiers in an ensemble system is introduced. Here we exploit the concept of information granule to construct granular prototypes for each class on the outputs of an ensemble of base classifiers. In the proposed method, uncertainty in the outputs of the base classifiers on training observations is captured by an interval-based representation. To predict the class label for a new observation, we first determine the distances between the output of the base classifiers for this observation and the class prototypes, then the predicted class label is obtained by choosing the label associated with the shortest distance. In the experimental study, we combine several learning algorithms to build the ensemble system and conduct experiments on the UCI, colon cancer, and selected CLEF2009 datasets. The experimental results demonstrate that the proposed framework outperforms several benchmarked algorithms including two trainable combining methods, i.e.

show abstract

Section: Discussionmentioning

confidence: 99%

“…Heterogeneous ensemble: Several different learning algorithms are learned on the same training set to generate the different base classifiers. The heterogeneous ensemble focuses more on the combining strategies on the meta-data [3,18,[23][24][25][26]) to achieve higher accuracy than a single classifier.…”

Section: Ensemble Methodsmentioning

confidence: 99%

Combining heterogeneous classifiers via granular prototypes

Nguyen

Pham

et al. 2018

Applied Soft Computing

Self Cite

View full text Add to dashboard Cite

show abstract

“…Optimization-based methods formulate ensemble selection as an optimization problem which can be solved by heuristic optimization or mathematical programming. In [22], Nguyen performance. In [5], Li et al theoretically analyzed the effect of diversity on the performance of voting.…”

Section: Ensemble Selection Methodsmentioning

confidence: 99%

Ensemble Selection based on Classifier Prediction Confidence

Nguyen

Luong

Dang

et al. 2020

Pattern Recognition

Self Cite

View full text Add to dashboard Cite

This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

show abstract

“…For instance, in [9] is proposed the utilization of multi-objective genetic algorithm (MOGA) in machine learning to improve generalization, learning and optimization abilities. Also focusing on genetic algorithms, in [10] is introduced an approach for classifier and feature selection in a multi-classifier system. In [11], a novel classification algorithm is proposed for a specific application of classifying thermostable protein by using Hurst exponent and SVM.…”

Section: Related Work and Brief Overviewmentioning

confidence: 99%

Electric field intensity for nonlinear classifier: A novel approach

Mendonca

Pinto

Marcato

2015

2015 IEEE 24th International Symposium on Industrial Electronics (ISIE)

View full text Add to dashboard Cite

-The classification through specific characteristics is of great relevance for development of machine intelligence systems in order to improve decision making capability. Spite of the existence of several classifier algorithms, they all have its drawbacks and might not be suitable for a certain application. Within the supervised learning classifiers, the definition of some parameters is a common issue, such as the learning step, number of hidden neurons in a multilayer perceptron, the margin length for the support vector machines, the overfitting in nonlinear classifiers, among other situations that in some way depend on a subjective decision. In order to overcome these issues, it is proposed in the present work a novel approach for a supervised classifier, which is capable of working with linear and nonlinear situations by calculating a hyperplane based on electric field intensity. To validate the method's performance, a well-known classification problem found in pattern recognition literature is tested and compared with another classical method. The obtained results have proven its good performance.

show abstract

A novel genetic algorithm approach for simultaneous feature and classifier selection in multi classifier system

Cited by 30 publications

References 15 publications

Combining heterogeneous classifiers via granular prototypes

Combining heterogeneous classifiers via granular prototypes

Ensemble Selection based on Classifier Prediction Confidence

Electric field intensity for nonlinear classifier: A novel approach

Contact Info

Product

Resources

About