Minimising the delta test for variable selection in regression problems

Guillén, Alberto; Sovilj, Dušan; Lendasse, Amaury; Mateo, Fernando; Rojas, Ignacio

doi:10.1504/ijhpsa.2008.024211

Cited by 36 publications

(47 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The crossover and mutation operators have also been implemented as custom functions according to [37]. The mutation operator is a pure random uniform function that operates at a gene level.…”

Section: Methodsmentioning

confidence: 99%

“…This value appears to be a good compromise between performance and computational cost for GA-based search applied to similarly sized datasets [37,58]. After some preliminary tests, the number of generations was fixed to 200 to ensure convergence in all cases.…”

Section: Methodsmentioning

confidence: 99%

“…To date, the GA has been successfully applied for variable selection in many publications [37,46,47,48,49,50]. Their success stems from the fact that they manage to carry out a global optimization of the selected set of variables.…”

Section: Using Genetic Algorithms For Global Searchmentioning

confidence: 99%

“…In the existing variable selection literature there are several applications of scaling to minimize the DT, but often keeping a discrete number of weights [36,37,38] instead of using unconstrained real values like in this study. In each generation of the GA, every input sample…”

Section: Scaling (S)mentioning

confidence: 99%

“…Several studies related to feature selection using the DT have been developed using both local search strategies such as forward search (FS) [36] and forward-backward selection (FBS) [37,38] and also global search techniques like tabu search [37,39] and Genetic Algorithm (GA) [37]. Global search methods have the advantage of being able to escape from local minima, to which local methods are prone to converge [40].…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Approximate k-NN delta test minimization method using genetic algorithms: Application to time series

2010

Self Cite

View full text Add to dashboard Cite

In many real world problems, the existence of irrelevant input variables (features) hinders the predictive quality of the models used to estimate the output variables. In particular, time series prediction often involves building large regressors of artificial variables that can contain irrelevant or misleading information. Many techniques have arisen to confront the problem of accurate variable selection, including both local and global search strategies. This paper presents a method based on genetic algorithms that intends to find a global optimum set of input variables that minimize the Delta Test criterion. The execution speed has been enhanced by substituting the exact nearest neighbor computation by its approximate version. The problems of scaling and projection of variables have been addressed. The developed method works in conjunction with MATLAB's Genetic Algorithm and Direct Search Toolbox. The goodness of the proposed methodology has been evaluated on several popular time series examples, and also generalized to other non-time-series datasets.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Methodsmentioning

confidence: 99%

Section: Using Genetic Algorithms For Global Searchmentioning

confidence: 99%

Section: Scaling (S)mentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Approximate k-NN delta test minimization method using genetic algorithms: Application to time series

2010

Self Cite

View full text Add to dashboard Cite

show abstract

Decision Support System to Determine Intention to Use Mobile Payment Systems on Social Networks: A Methodological Analysis

Guillén

Herrera

Pomares

et al. 2015

Int. J. Intell. Syst.

Self Cite

View full text Add to dashboard Cite

The rapid growth of social networks has led many companies to use mobile payment systems as business sales tools. As these platforms have an increasing acceptance among the consumers, the main goal of this research is to analyze the individuals' use intention of these systems in a social network environment. The data used were collected after performing an experiment using the Zong Payment tool. The paper includes a comparative analysis of several machine learning methods applied to the variable selection problem considering strategic, sociodemographic, and behavioral variables. The methods were applied isolatedly and, in a novel approach, using multiobjective optimization criteria. The results obtained by these algorithms were validated by an expert committee, who ranked them on the basis of their quality. The proposed methodology obtained the best results in terms of optimization as well as the highest scores from the experts. C 2015 Wiley Periodicals, Inc.

show abstract