Equilibrium Optimizer for Feature Selection

Introduction

• This toolbox offers an Equilibrium Optimizer ( EO ) method
• The Main file illustrates the example of how EO can solve the feature selection problem using benchmark data-set.

Input

• feat : feature vector ( Instances x Features )
• label : label vector ( Instances x 1 )
• N : number of particles
• max_Iter : maximum number of iterations
• a1 : Parameter
• a2 : Parameter
• GP : Generation rate control parameter

Output

• sFeat : selected features
• Sf : selected feature index
• Nf : number of selected features
• curve : convergence curve

Example

% Benchmark data set 
load ionosphere.mat; 

% Set 20% data as validation set
ho = 0.2; 
% Hold-out method
HO = cvpartition(label,'HoldOut',ho);

% Parameter setting
N        = 10; 
max_Iter = 100;
a1       = 2;      % constant
a2       = 1;      % constant
GP       = 0.5;    % generation probability 

% Equilibrium Optimizer
[sFeat,Sf,Nf,curve] = jEO(feat,label,N,max_Iter,a1,a2,GP,HO);

% Plot convergence curve
plot(1:max_Iter,curve);
xlabel('Number of iterations');
ylabel('Fitness Value');
title('EO'); grid on;

Requirement

• MATLAB 2014 or above
• Statistics and Machine Learning Toolbox

Cite As

@article{too2020general,
  title={General Learning Equilibrium Optimizer: A New Feature Selection Method for Biological Data Classification},
  author={Too, Jingwei and Mirjalili, Seyedali},
  journal={Applied Artificial Intelligence},
  pages={1--17},
  year={2020},
  publisher={Taylor \& Francis}
}