Skip to main content
Fig. 2 | BMC Bioinformatics

Fig. 2

From: In silico prediction of potential chemical reactions mediated by human enzymes

Fig. 2

Performances (AUC) of four machine learning algorithms with the increasing number of features. The four machine learning algorithms are artificial neural network, random forest, naïve Bayes and multiple linear regression. Their performances were calculated by 20-fold cross-validation. Of the four algorithms, multiple linear regression model using 500 features showed the best performance (AUC = 0.897)

Back to article page