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

Accuracy: 0.352

Confusion matrix
Real	Pred.	#
N	NONE	198
N	N	196
N	P	40
N	NEU	38
NEU	NEU	31
NEU	NONE	192
NEU	N	91
NEU	P	53
NONE	NONE	88
NONE	N	37
NONE	P	23
NONE	NEU	11
P	N	59
P	P	187
P	NONE	158
P	NEU	26

Category evaluation
Cat	P	R	F1
N	0.512	0.415	0.458
NEU	0.292	0.084	0.131
NONE	0.138	0.553	0.221
P	0.617	0.435	0.510

Macro-Precision: 0.390
Macro-Recall: 0.372
Macro-F1: 0.381