Skip to main content

Table 4 A grid of parameter values for different classifiers

From: Comparison of six machine learning methods for differentiating benign and malignant thyroid nodules using ultrasonographic characteristics

Classifier

Parameter Values

SVM

c ∈ {− 8: 0.8: 8}; g ∈ {− 8: 0.8: 8}

RF

m ∈ {1: 1: 13}; ntree = 500

GlmNet

α ∈ {0.1: 0.1: 1}; λ ∈ {2.0789e − 4, 3.6329e − 4, 6.3485e − 04, 0.0011, 0.0019, 0.0034, 0.059, 0.0103}

K-NN

k ∈ {2: 2: 22}

Back to article page

BMC Medical Imaging

ISSN: 1471-2342

Contact us