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

Accuracy: 0.530

Confusion matrix
Real	Pred.	#
N	NEU	86
N	N	303
N	P	38
N	NONE	64
NEU	N	57
NEU	NEU	34
NEU	P	33
NEU	NONE	40
NONE	N	56
NONE	NONE	99
NONE	NEU	36
NONE	P	33
P	P	217
P	N	40
P	NONE	47
P	NEU	50

Category evaluation
Cat	P	R	F1
N	0.664	0.617	0.640
NEU	0.165	0.207	0.184
NONE	0.396	0.442	0.418
P	0.676	0.613	0.643

Macro-Precision: 0.475
Macro-Recall: 0.470
Macro-F1: 0.473