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

Accuracy: 0.523

Confusion matrix
Real	Pred.	#
N	N	414
N	P	45
N	NEU	6
N	NONE	7
NEU	N	211
NEU	NEU	17
NEU	NONE	31
NEU	P	108
NONE	N	74
NONE	NONE	19
NONE	P	57
NONE	NEU	9
P	N	113
P	P	297
P	NONE	7
P	NEU	13

Category evaluation
Cat	P	R	F1
N	0.510	0.877	0.645
NEU	0.378	0.046	0.083
NONE	0.297	0.119	0.170
P	0.586	0.691	0.634

Macro-Precision: 0.443
Macro-Recall: 0.433
Macro-F1: 0.438