TABLE 2.

Dietary biomarkers document classification results. Highest precision, recall, F1-score, F2-score and AUC achieved by each algorithm: Decision Tree (DT), Logistic Regression (LR), Naïve Bayes (NB), Neural Network (NN), Random Forest (RF) and Support Vector Machine (SVM). The highest value of each metric on each feature type is bolded.

TITLES
	MaxPrecision	MaxRecall	MaxF1	MaxF2	Max-AUC
DT	0.216	0.433	0.262	0.302	0.635
LR	0.388	0.707	0.495	0.601	0.910
NB	0.528	0.652	0.475	0.561	0.903
NN	0.560	0.331	0.385	0.348	0.887
RF	0.415	0.661	0.489	0.577	0.889
SVM	0.586	0.688	0.462	0.560	0.904

ABSTRACTS
	MaxPrecision	MaxRecall	MaxF1	MaxF2	Max-AUC
DT	0.337	0.570	0.397	0.449	0.720
LR	0.559	0.770	0.618	0.687	0.953
NB	0.606	0.752	0.644	0.684	0.952
NN	0.854	0.441	0.542	0.472	0.948
RF	0.621	0.705	0.591	0.644	0.954
SVM	0.512	0.798	0.572	0.658	0.949

TITLES + ABSTRACTS
	MaxPrecision	MaxRecall	MaxF1	MaxF2	Max-AUC
DT	0.379	0.533	0.419	0.480	0.740
LR	0.550	0.779	0.614	0.701	0.948
NB	0.601	0.788	0.643	0.695	0.950
NN	0.764	0.432	0.528	0.463	0.945
RF	0.665	0.687	0.586	0.634	0.953
SVM	0.512	0.807	0.564	0.664	0.948

TITLES + METADATA
	MaxPrecision	MaxRecall	MaxF1	MaxF2	Max-AUC
DT	0.307	0.468	0.328	0.390	0.693
LR	0.392	0.734	0.502	0.616	0.930
NB	0.373	0.707	0.456	0.566	0.914
NN	0.603	0.266	0.333	0.274	0.905
RF	0.476	0.725	0.492	0.592	0.924
SVM	0.163	0.168	0.144	0.154	0.725

The highest value of each metric is bolded.