A Lightweight API-Based Approach for Building Flexible Clinical NLP Systems

Table 3

Impact of negation from the experiments.

Dataset	Negation (κ)	Recall	Precision	F1 score
Obesity challenge	True	0.733	0.939	0.823
	False	0.805	0.925	0.861
Medication challenge	True	0.62	0.835	0.712
	False	0.636	0.838	0.724
OPERAM medical conditions	True	0.594	0.271	0.373
	False	0.594	0.271	0.373
OPERAM medications	True	0.795	0.816	0.805
	False	0.795	0.816	0.805