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

Accuracy: 0.472

Confusion matrix
Real	Pred.	#
N	P	46
N	N	261
N	NONE	61
N	NEU	104
NEU	N	105
NEU	NEU	75
NEU	NONE	99
NEU	P	88
NONE	NONE	46
NONE	N	36
NONE	P	46
NONE	NEU	31
P	P	292
P	NEU	73
P	NONE	31
P	N	34

Category evaluation
Cat	P	R	F1
N	0.599	0.553	0.575
NEU	0.265	0.204	0.231
NONE	0.194	0.289	0.232
P	0.619	0.679	0.647

Macro-Precision: 0.419
Macro-Recall: 0.431
Macro-F1: 0.425