Supporting the classification of patients in public hospitals in Chile by designing, deploying and validating a system based on natural language processing

Table 4 Performance of the Random Forest Classifier over the testing dataset and the ground truth constructed from human classifications

Class	Precision	Recall	F1-score	Number of examples
Testing dataset
GES	0.67	0.90	0.77	37,502
non-GES	0.98	0.91	0.94	173,011
Weighted average	0.92	0.91	0.91	210,513
Ground truth dataset
GES	0.92	0.55	0.69	260
No-GES	0.85	0.98	0.91	681
Weighted average	0.87	0.86	0.85	941

ISSN: 1472-6947