Skip to main content

Table 3 Parameters settings

From: Application of genetic algorithms and constructive neural networks for the analysis of microarray cancer data

Algorithm

Test Parameters

LDA

No parameters

SVM

Kernel type, t= {linear, polynomial, radial base function, sigmoid}

Cost, C = {1, 3, 5, 7, 9, 10, 12, 15}

Degree, d = {1, 2, 3, 4, 5}

Gamma, g = {0.001, 0.005, 0.1, 0.15, 0.2, 0.4, 0.6, 0.8, 1, 2, 3, 5}

Coef0, r= {0, 1, 2}

NaiveBayes

Kernel density, K = {0, 1}

Supervised discretization, D = {0, 1}

C-MANTEC

Max. Iterations, I max = {1000, 10000, 100000}

GFac, g fac = {0.01, 0.05, 0.1, 0.2, 0.25, 0.3}

Phi, φ = {1, 1.5, 2, 2.5, 3, 3.5, 4, 4.5, 5, 5.5, 6}

kNN

Neighbours, k = {1, 2, 3, 4, 5, 6, 7, 8, 9, 10}

Distance type, d= {Euclidean, chi-squared, cosine-similarity}

MLP

Hidden neurons, N Hidden = {2, 3, 4, 5, 6}

Alpha, α = {0.05, 0.1, 0.2, 0.3, 0.5}

Number cycles, N Cycles = {10, 25, 50}

  1. Parameter settings tested during evaluation of the classification algorithms. The combination of all the values of the parameters generate a set of configurations for each method.