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

Accuracy: 0.390

Confusion matrix
Real	Pred.	#
N	N	228
N	NONE	233
N	P	30
NEU	N	45
NEU	NONE	97
NEU	P	22
NONE	NONE	174
NONE	N	29
NONE	P	21
P	NONE	225
P	N	50
P	P	79

Category evaluation
Cat	P	R	F1
N	0.648	0.464	0.541
NEU	0.000	0.000	0.000
NONE	0.239	0.777	0.365
P	0.520	0.223	0.312

Macro-Precision: 0.352
Macro-Recall: 0.366
Macro-F1: 0.359