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

Accuracy: 0.366

Confusion matrix
Real	Pred.	#
N	NONE	167
N	N	223
N	NEU	59
N	P	23
NEU	NONE	187
NEU	N	96
NEU	NEU	54
NEU	P	30
NONE	NONE	86
NONE	NEU	15
NONE	P	16
NONE	N	42
P	N	51
P	P	160
P	NONE	164
P	NEU	55

Category evaluation
Cat	P	R	F1
N	0.541	0.472	0.505
NEU	0.295	0.147	0.196
NONE	0.142	0.541	0.225
P	0.699	0.372	0.486

Macro-Precision: 0.419
Macro-Recall: 0.383
Macro-F1: 0.400