Evaluation - task1-intertass-ST4-CR
1233 tweets in qrel, 1233 tweets in run.

Accuracy: 0.538

Confusion matrix
Real	Pred.	#
N	N	335
N	NEU	61
N	P	53
N	NONE	42
NEU	N	73
NEU	NONE	23
NEU	P	46
NEU	NEU	22
NONE	NEU	34
NONE	NONE	77
NONE	P	49
NONE	N	64
P	P	229
P	N	54
P	NEU	38
P	NONE	33

Category evaluation
Cat	P	R	F1
N	0.637	0.682	0.659
NEU	0.142	0.134	0.138
NONE	0.440	0.344	0.386
P	0.607	0.647	0.627

Macro-Precision: 0.457
Macro-Recall: 0.452
Macro-F1: 0.454