1 模型
2 部分代码
%-------------------------------------------------------------------------%
% Binary Dragonfly Algorithm (BDA) source codes demo version % %
%-------------------------------------------------------------------------%
%---Inputs-----------------------------------------------------------------
% feat: features
% label: labelling
% N: Number of dragonflies
% T: Maximum number of iterations
% Dmax: Maximum velocity
% *Note: k-value of KNN & hold-out setting can be modified in jFitnessFunction.m
%---Outputs----------------------------------------------------------------
% sFeat: Selected features
% Sf: Selected feature index
% Nf: Number of selected features
% curve: Convergence curve
%--------------------------------------------------------------------------
%% Binary Dragonfly Algorithm
clc, clear, close;
% Benchmark data set
load ionosphere.mat;
% Parameter setting
N=10;
T=100;
Dmax=6;
% Binary Dragonfly Algorithm
[sFeat,Sf,Nf,curve]=jBDA(feat,label,N,T,Dmax);
% Plot convergence curve
figure(); plot(1:T,curve); xlabel('Number of iterations');
ylabel('Fitness Value'); title('BDA'); grid on;
3 仿真结果
4 参考文献
[1]董海, 徐德珉. 基于蜻蜓算法和最小二乘向量机的小批量生产质量预测[J]. 科技管理研究, 2019, 000(022):256-260.