Evaluation - task1-intertass-ES
1899 tweets in qrel, 500 tweets in run.

Accuracy: 0.096

Confusion matrix
Real	Pred.	#
N	N	119
N	P	71
N	NEU	6
N	NONE	14
N	?	557
NEU	N	38
NEU	P	28
NEU	NONE	3
NEU	?	147
NONE	N	36
NONE	P	24
NONE	NONE	1
NONE	?	213
P	N	86
P	NONE	9
P	P	62
P	NEU	3
P	?	482

Category evaluation
Cat	P	R	F1
N	0.427	0.155	0.228
NEU	0.000	0.000	0.000
NONE	0.037	0.004	0.007
P	0.335	0.097	0.150

Macro-Precision: 0.200
Macro-Recall: 0.064
Macro-F1: 0.097