Evaluation - task1-intertass-ST4-PE
1428 tweets in qrel, 1428 tweets in run.

Accuracy: 0.514

Confusion matrix
Real	Pred.	#
N	P	67
N	N	372
N	NEU	1
N	NONE	32
NEU	P	118
NEU	N	182
NEU	NONE	62
NEU	NEU	5
NONE	NONE	43
NONE	NEU	5
NONE	P	52
NONE	N	59
P	P	314
P	N	98
P	NONE	16
P	NEU	2

Category evaluation
Cat	P	R	F1
N	0.523	0.788	0.629
NEU	0.385	0.014	0.026
NONE	0.281	0.270	0.276
P	0.570	0.730	0.640

Macro-Precision: 0.440
Macro-Recall: 0.451
Macro-F1: 0.445