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

Accuracy: 0.468

Confusion matrix
Real	Pred.	#
N	NEU	33
N	NONE	193
N	N	227
N	P	38
NEU	N	40
NEU	NEU	20
NEU	P	34
NEU	NONE	70
NONE	N	41
NONE	NEU	9
NONE	NONE	131
NONE	P	43
P	P	199
P	NEU	13
P	NONE	107
P	N	35

Category evaluation
Cat	P	R	F1
N	0.662	0.462	0.544
NEU	0.267	0.122	0.167
NONE	0.261	0.585	0.361
P	0.634	0.562	0.596

Macro-Precision: 0.456
Macro-Recall: 0.433
Macro-F1: 0.444