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

Accuracy: 0.317

Confusion matrix
Real	Pred.	#
N	P	27
N	NONE	269
N	N	176
NEU	NONE	269
NEU	N	71
NEU	P	27
NONE	NONE	116
NONE	N	20
NONE	P	22
NONE	NEU	1
P	P	160
P	NONE	237
P	N	31
P	NEU	2

Category evaluation
Cat	P	R	F1
N	0.591	0.373	0.457
NEU	0.000	0.000	0.000
NONE	0.130	0.730	0.221
P	0.678	0.372	0.480

Macro-Precision: 0.350
Macro-Recall: 0.369
Macro-F1: 0.359