Machine learning analysis of microbial flow cytometry data from nanoparticles, antibiotics and carbon sources perturbed anaerobic microbiomes

Table 2 Area under the curve (AUC) values corresponding to Receiver Operating Characteristics (ROC) curves shown in Fig. 2 for test data (^* Demonstrated deep learning model was a feed forward artificial neural network with three hidden layers)

Putative Groups	Gradient Boosting	Naïve Bayes	Distributed Random Forests	Deep Learning^*
Acetogens	0.7829	0.7279	0.6482	0.7853
Acidogens	0.9993	0.9999	0.9833	0.9983
Hydrolyzers	0.9638	0.9391	0.8055	0.9269
Methanogens	0.8520	0.8024	0.7773	0.8585

ISSN: 1754-1611