Skip to main content

Table 4 Mean testing accuracy of individual classification models after 50 replicates with standard deviation in the brackets

From: Ensemble machine learning approach for screening of coronary heart disease based on echocardiography and risk factors

Model

Accuracy

logistic regression

\(67.7\% (0.034)\)

penalized logistic regression

\(70.8\% (0.022)\)

cumulative probability model

\(68.6\% (0.035)\)

random forest

\(59.2\% (0.034)\)

weighted subspace random forest

\(59.3\% (0.033)\)

SVM with class weight

\(70.2\% (0.043)\)

SVM with polynomial kernel

\(66.3\% (0.041)\)

SVM with radial kernel

\(63.7\% (0.041)\)

K-nearest neighbor

\(58.2\% (0.037)\)

LDA

\(69.6\% (0.048)\)

sparsed LDA

\(58.8\% (0.036)\)

naive Bayes

\(64.4\% (0.024)\)

Bayes generalized linear model

\(68.0\% (0.031)\)

Gaussian process with polynomial kernel

\(70.1\% (0.035)\)

Gaussian process with radial kernel

\(65.2\% (0.029)\)

Neural network

\(62.8\% (0.043)\)

Monotone multi-layer perceptron neural network

\(69.2\% (0.026)\)

model average neural network

\(65.1\% (0.035)\)

stochastic gradient boosting

\(57.8\% (0.027)\)