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

Accuracy: 0.438

Confusion matrix
Real	Pred.	#
N	P	124
N	N	203
N	NONE	83
N	NEU	62
NEU	NONE	74
NEU	N	84
NEU	P	160
NEU	NEU	49
NONE	NONE	40
NONE	N	31
NONE	P	71
NONE	NEU	17
P	P	333
P	N	40
P	NEU	31
P	NONE	26

Category evaluation
Cat	P	R	F1
N	0.567	0.430	0.489
NEU	0.308	0.134	0.186
NONE	0.179	0.252	0.209
P	0.484	0.774	0.596

Macro-Precision: 0.385
Macro-Recall: 0.397
Macro-F1: 0.391