Evaluation - task1-intertass-ST4-ES
1899 tweets in qrel, 1899 tweets in run.

Accuracy: 0.466

Confusion matrix
Real	Pred.	#
N	NEU	217
N	N	279
N	NONE	208
N	P	63
NEU	NONE	80
NEU	NEU	72
NEU	P	34
NEU	N	30
NONE	NONE	147
NONE	P	49
NONE	NEU	55
NONE	N	23
P	NONE	123
P	P	387
P	N	30
P	NEU	102

Category evaluation
Cat	P	R	F1
N	0.771	0.364	0.494
NEU	0.161	0.333	0.218
NONE	0.263	0.536	0.353
P	0.726	0.603	0.659

Macro-Precision: 0.480
Macro-Recall: 0.459
Macro-F1: 0.470