Skip to main content
. 2023 Dec 22;31(3):692–704. doi: 10.1093/jamia/ocad244

Figure 2.

Figure 2.

Negative log loss (smaller values are better) ML performance of representative examples of combinations of classifiers and feature sets evaluated by applying cross-validation on the training set. Abbreviations: NGRAM = uni- and bigram sentence features, BERT = ClinicalBert 768 dimension embedding vector, LOGISTIC_REGRESSION = logistic regression classifier in scikit-learn, default parameters, NN_MLP_1024 = neural network MLPClassifier in scikit-learn with hidden layer of size 1024, RANDOM_FOREST_200 = RandomForest classifier in scikit-learn with 200 estimators, other parameters at defaults, LINEAR_SVC_C_1.0 = support vector machine classifier, scikit-lean SVC implementation with C parameter set to 1.0.