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

Accuracy: 0.459

Confusion matrix
Real	Pred.	#
N	P	133
N	NEU	90
N	N	246
N	NONE	3
NEU	P	152
NEU	NEU	101
NEU	N	112
NEU	NONE	2
NONE	N	45
NONE	NEU	36
NONE	P	78
P	P	309
P	NEU	44
P	N	77

Category evaluation
Cat	P	R	F1
N	0.513	0.521	0.517
NEU	0.373	0.275	0.317
NONE	0.000	0.000	0.000
P	0.460	0.719	0.561

Macro-Precision: 0.336
Macro-Recall: 0.379
Macro-F1: 0.356