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

Accuracy: 0.375

Confusion matrix
Real	Pred.	#
N	P	61
N	NONE	165
N	N	221
N	NEU	25
NEU	NONE	204
NEU	NEU	25
NEU	P	53
NEU	N	85
NONE	NONE	88
NONE	NEU	10
NONE	N	31
NONE	P	30
P	P	202
P	NONE	153
P	N	56
P	NEU	19

Category evaluation
Cat	P	R	F1
N	0.562	0.468	0.511
NEU	0.316	0.068	0.112
NONE	0.144	0.553	0.229
P	0.584	0.470	0.521

Macro-Precision: 0.402
Macro-Recall: 0.390
Macro-F1: 0.396