Detecting Personal Medication Intake in Twitter via Domain Attention-Based RNN with Multi-Level Features

Table 2

Comparison results.

Model	Precision	Recall	F1
NB	0.675	0.631	0.650
SVM	0.679	0.664	0.668
BiLSTM	0.683	0.672	0.678
CharCNN	0.681	0.697	0.689
AttRNN	0.704	0.677	0.688
InfyNLP	0.725^#	0.664^#	0.693^#
UKNLP	0.701^#	0.677^#	0.689^#
NRC-Canada	0.704^#	0.635^#	0.668^#
Our model	0.708	0.694	0.697