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

Accuracy: 0.461

Confusion matrix
Real	Pred.	#
N	NEU	126
N	P	41
N	N	223
N	NONE	82
NEU	NEU	105
NEU	N	64
NEU	NONE	117
NEU	P	81
NONE	NONE	56
NONE	P	44
NONE	N	32
NONE	NEU	27
P	N	37
P	NONE	37
P	P	275
P	NEU	81

Category evaluation
Cat	P	R	F1
N	0.626	0.472	0.539
NEU	0.310	0.286	0.297
NONE	0.192	0.352	0.248
P	0.624	0.640	0.631

Macro-Precision: 0.438
Macro-Recall: 0.438
Macro-F1: 0.438