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

Accuracy: 0.426

Confusion matrix
Real	Pred.	#
N	NONE	132
N	P	59
N	NEU	77
N	N	204
NEU	NONE	139
NEU	NEU	81
NEU	N	69
NEU	P	78
NONE	NONE	75
NONE	P	40
NONE	N	28
NONE	NEU	16
P	P	249
P	NEU	73
P	NONE	68
P	N	40

Category evaluation
Cat	P	R	F1
N	0.598	0.432	0.502
NEU	0.328	0.221	0.264
NONE	0.181	0.472	0.262
P	0.585	0.579	0.582

Macro-Precision: 0.423
Macro-Recall: 0.426
Macro-F1: 0.424