Evaluation - task1-intertass-ST4-PE
1428 tweets in qrel, 1428 tweets in run.

Accuracy: 0.505

Confusion matrix
Real	Pred.	#
N	P	76
N	N	348
N	NONE	45
N	NEU	3
NEU	P	113
NEU	N	154
NEU	NONE	93
NEU	NEU	7
NONE	NONE	52
NONE	N	56
NONE	P	48
NONE	NEU	3
P	P	314
P	N	84
P	NONE	27
P	NEU	5

Category evaluation
Cat	P	R	F1
N	0.542	0.737	0.625
NEU	0.389	0.019	0.036
NONE	0.240	0.327	0.277
P	0.570	0.730	0.640

Macro-Precision: 0.435
Macro-Recall: 0.453
Macro-F1: 0.444