Evaluation - task1-intertass-PE
1428 tweets in qrel, 1428 tweets in run.

Accuracy: 0.382

Confusion matrix
Real	Pred.	#
N	N	185
N	NEU	180
N	P	44
N	NONE	63
NEU	N	103
NEU	NEU	128
NEU	NONE	79
NEU	P	57
NONE	NEU	54
NONE	NONE	37
NONE	P	26
NONE	N	42
P	N	82
P	NEU	101
P	P	195
P	NONE	52

Category evaluation
Cat	P	R	F1
N	0.449	0.392	0.419
NEU	0.276	0.349	0.308
NONE	0.160	0.233	0.190
P	0.606	0.453	0.519

Macro-Precision: 0.373
Macro-Recall: 0.357
Macro-F1: 0.365