Table 1 Prediction models
From: Ensemble of rankers for efficient gene signature extraction in smoke exposure classification
Classifier | Acronym | Parameters |
|---|---|---|
Random forests | RF | split=gini, max depth=none, min samples leaf=1, min samples split=1, max features=auto, no. estimators=10 |
Gaussian Naive Bayes | GNB | none |
k–Nearest neighbors | kNN | no.neighbors=3, algorithm=auto, metric=minkowski, p=2, weights=uniform, leaf size=30 |
MultiLayer perceptron | MLP | activation=relu;algorithm=l-bfgs, α=1e-05, β1=0.9, beta2=0.999, ε=1e-08, hidden layer sizes=(100,) |
Support vector classifier | SVC | kernel=linear, C=0.1, tolerance=0.001 |
Logistic regression | LR | C=1.0 max iter=100 penalty=L2 tolerance=0.0001, multi class=OvR |
Linear discriminant analysis | LDA | solver=SVD, tolerance=0.0001 |
Gradient tree boosting | GTB | loos=deviance, subsample=1.0 learning rate=0.1, min sample split=2, mean sample leaf=1, max depth=3, estimators=100 |
Extremely randomized trees | ERT | split=gini, max depth=No, min samples leaf=1, min samples split=1, max features=auto, no. estimators=10 |