Evaluation - task1-intertass-ES
1899 tweets in qrel, 1008 tweets in run.

Accuracy: 0.191

Confusion matrix
Real	Pred.	#
N	N	251
N	P	131
N	NONE	14
N	NEU	1
N	?	370
NEU	N	75
NEU	NONE	5
NEU	P	43
NEU	?	93
NONE	N	93
NONE	P	47
NONE	NONE	7
NONE	NEU	1
NONE	?	126
P	N	211
P	P	104
P	NONE	22
P	NEU	3
P	?	302

Category evaluation
Cat	P	R	F1
N	0.398	0.327	0.359
NEU	0.000	0.000	0.000
NONE	0.146	0.026	0.043
P	0.320	0.162	0.215

Macro-Precision: 0.216
Macro-Recall: 0.129
Macro-F1: 0.161