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

Accuracy: 0.422

Confusion matrix
Real	Pred.	#
N	N	284
N	P	137
N	NONE	30
N	NEU	21
NEU	NEU	15
NEU	N	170
NEU	P	146
NEU	NONE	36
NONE	P	67
NONE	N	64
NONE	NONE	22
NONE	NEU	6
P	N	115
P	P	281
P	NONE	26
P	NEU	8

Category evaluation
Cat	P	R	F1
N	0.449	0.602	0.514
NEU	0.300	0.041	0.072
NONE	0.193	0.138	0.161
P	0.445	0.653	0.530

Macro-Precision: 0.347
Macro-Recall: 0.359
Macro-F1: 0.353