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

Accuracy: 0.405

Confusion matrix
Real	Pred.	#
N	N	250
N	NONE	185
N	P	37
NEU	N	110
NEU	NONE	198
NEU	P	59
NONE	NONE	90
NONE	N	42
NONE	P	27
P	P	238
P	NONE	141
P	N	51

Category evaluation
Cat	P	R	F1
N	0.552	0.530	0.541
NEU	0.000	0.000	0.000
NONE	0.147	0.566	0.233
P	0.659	0.553	0.602

Macro-Precision: 0.339
Macro-Recall: 0.412
Macro-F1: 0.372