Evaluation - task1-intertass-ST4-CR
1233 tweets in qrel, 1233 tweets in run.

Accuracy: 0.423

Confusion matrix
Real	Pred.	#
N	N	195
N	NONE	203
N	NEU	58
N	P	35
NEU	N	18
NEU	NONE	110
NEU	P	23
NEU	NEU	13
NONE	N	31
NONE	NONE	155
NONE	P	20
NONE	NEU	18
P	P	159
P	N	30
P	NEU	18
P	NONE	147

Category evaluation
Cat	P	R	F1
N	0.712	0.397	0.510
NEU	0.121	0.079	0.096
NONE	0.252	0.692	0.369
P	0.671	0.449	0.538

Macro-Precision: 0.439
Macro-Recall: 0.404
Macro-F1: 0.421