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

Accuracy: 0.395

Confusion matrix
Real	Pred.	#
N	NEU	137
N	P	301
N	N	278
N	NONE	51
NEU	NEU	33
NEU	N	60
NEU	P	108
NEU	NONE	15
NONE	P	139
NONE	N	66
NONE	NEU	39
NONE	NONE	30
P	P	409
P	N	113
P	NEU	78
P	NONE	42

Category evaluation
Cat	P	R	F1
N	0.538	0.362	0.433
NEU	0.115	0.153	0.131
NONE	0.217	0.109	0.146
P	0.427	0.637	0.512

Macro-Precision: 0.324
Macro-Recall: 0.315
Macro-F1: 0.320