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

Accuracy: 0.420

Confusion matrix
Real	Pred.	#
N	P	33
N	NONE	131
N	N	176
N	NEU	132
NEU	P	72
NEU	NONE	122
NEU	N	72
NEU	NEU	101
NONE	NONE	64
NONE	N	23
NONE	P	40
NONE	NEU	32
P	P	259
P	NEU	83
P	N	22
P	NONE	66

Category evaluation
Cat	P	R	F1
N	0.601	0.373	0.460
NEU	0.290	0.275	0.283
NONE	0.167	0.403	0.236
P	0.641	0.602	0.621

Macro-Precision: 0.425
Macro-Recall: 0.413
Macro-F1: 0.419