Optimization of stacking ensemble configurations through Artificial Bee Colony algorithm

Shunmugapriya, P.; Kanmani, S.

doi:10.1016/j.swevo.2013.04.004

Cited by 52 publications

(37 citation statements)

References 39 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…So, if the quadratic model is selected for both dataset, it can perform better than the other model. This model can be stated by following equations, for Glass and Iris, respectively: 6 10 6139 9 10 5747 1 10 2881 2 10 6648 8 10 4694 4 10 6912 3 10 7455 9 10 3417 1 10 5905 2 10 3127 1 3333 0 0434 0 0923 0 3584 0 0274 0 0977 0 0817 0 0508 0 0842 0 4285 0 0291 0 1258 0 1088 0 According to above equations, for Glass, scale 2 , shift 1 and the square of scale 2 are respectively more important for ensemble size, error rate and diversity, because their related β is larger. For the same reason, shift 2 , c 2 and shift 2 have greater impact for Iris in ensemble size, error and diversity, respectively.…”

Section: Dataset Algorithm Ensemble Size Error Rate Q Statisticmentioning

confidence: 99%

See 1 more Smart Citation

Design and Stability Analysis of Multi-Objective Ensemble Classifiers

Pourtaheri¹,

Zahiri²,

Razavi³

2017

ELCVIA

View full text Add to dashboard Cite

In this paper a novel multi-objective optimization technique based on Inclined Planes Optimization algorithm (called MOIPO), is used to design an ensemble classifier. Diversity, ensemble size, and error rate are three objectives which are considered along designing the proposed ensemble classifier.The performance of designed ensemble classifier is tested on different kinds of benchmarks with nonlinear, overlapping class boundaries, and different feature space dimensions. Extensive experimental and comparative results on these data sets provided to show the performance of the proposed method, are better than ensemble designed by Multi-Objective Particle Swarm Optimization (MOPSO) algorithm.Another important aspect of this article is stability analysis of designed ensemble classifier. In fact, for the first time, the stability of a heuristic ensemble classifier is analyzed by using statistical method. For this aim, three regression models are investigated by applying F-test to find better model in each case. Due to the results of stability analysis, quadratic model is the best model for two datasets as representative of simple data and overlapped data.

show abstract

Section: Dataset Algorithm Ensemble Size Error Rate Q Statisticmentioning

confidence: 99%

“…In other words, an ensemble classifier combines a finite number of classifiers of the same type or different, trained concurrently for a joint classification problem. The ensemble efficiently amends the performance of the classifier compared to a single classifier [2].…”

Section: Introductionmentioning

confidence: 99%

Design and Stability Analysis of Multi-Objective Ensemble Classifiers

Pourtaheri¹,

Zahiri²,

Razavi³

2017

ELCVIA

View full text Add to dashboard Cite

show abstract

“…Since ABC has few control parameters and also easy to implement, it has been widely used in many optimization applications, such as dynamic anomaly detection in MANETs [15], stacking ensemble [16], classification rule discovery [50], and so on.…”

Section: Artificial Bee Colonymentioning

confidence: 99%

“…It is as simple as particle swarm optimization (PSO) [13] and differential evolution (DE) [14], and uses only common control parameters, such as population size and maximum cycle number. ABC has shown promising results in the field of optimization [15,16].…”

Section: Introductionmentioning

confidence: 99%

BeeOWA: A novel approach based on ABC algorithm and induced OWA operators for constructing one-class classifier ensembles

Parhizkar

Abadi

2015

Neurocomputing

View full text Add to dashboard Cite

“…There are also some new papers which propose extensions to SR (Džeroski and Ženko (2004), Rooney et al (2004a), Rooney et al (2004b), Xu et al (2007), Ozay and Vural (2008), Ni et al (2009)), Ledezma et al (2010), Shunmugapriya and Kanmani (2013). An informative overview of SR methods can be found in Sesmero et al (2015).…”

Section: Introductionmentioning

confidence: 99%

Stacked Regression With a Generalization of the Moore-Penrose Pseudoinverse

Górecki

Łuczak

2017

Statistics in Transition New Series

View full text Add to dashboard Cite

In practice, it often happens that there are a number of classification methods. We are not able to clearly determine which method is optimal. We propose a combined method that allows us to consolidate information from multiple sources in a better classifier. Stacked regression (SR) is a method for forming linear combinations of different classifiers to give improved classification accuracy. The Moore-Penrose (MP) pseudoinverse is a general way to find the solution to a system of linear equations.This paper presents the use of a generalization of the MP pseudoinverse of a matrix in SR. However, for data sets with a greater number of features our exact method is computationally too slow to achieve good results: we propose a genetic approach to solve the problem. Experimental results on various real data sets demonstrate that the improvements are efficient and that this approach outperforms the classical SR method, providing a significant reduction in the mean classification error rate.

show abstract

Optimization of stacking ensemble configurations through Artificial Bee Colony algorithm

Cited by 52 publications

References 39 publications

Design and Stability Analysis of Multi-Objective Ensemble Classifiers

Design and Stability Analysis of Multi-Objective Ensemble Classifiers

BeeOWA: A novel approach based on ABC algorithm and induced OWA operators for constructing one-class classifier ensembles

Stacked Regression With a Generalization of the Moore-Penrose Pseudoinverse

Contact Info

Product

Resources

About