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

Accuracy: 0.512

Confusion matrix
Real	Pred.	#
N	N	363
N	P	71
N	NONE	50
N	NEU	7
NEU	N	84
NEU	NEU	10
NEU	P	51
NEU	NONE	19
NONE	N	97
NONE	P	71
NONE	NEU	4
NONE	NONE	52
P	NONE	45
P	N	94
P	P	206
P	NEU	9

Category evaluation
Cat	P	R	F1
N	0.569	0.739	0.643
NEU	0.333	0.061	0.103
NONE	0.313	0.232	0.267
P	0.516	0.582	0.547

Macro-Precision: 0.433
Macro-Recall: 0.404
Macro-F1: 0.418