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

Accuracy: 0.520

Confusion matrix
Real	Pred.	#
N	N	310
N	NEU	42
N	P	76
N	NONE	63
NEU	N	57
NEU	P	58
NEU	NEU	17
NEU	NONE	32
NONE	N	51
NONE	NONE	92
NONE	NEU	18
NONE	P	63
P	P	222
P	N	50
P	NEU	30
P	NONE	52

Category evaluation
Cat	P	R	F1
N	0.662	0.631	0.647
NEU	0.159	0.104	0.125
NONE	0.385	0.411	0.397
P	0.530	0.627	0.574

Macro-Precision: 0.434
Macro-Recall: 0.443
Macro-F1: 0.439