Skip to main content

View full-text article in PMC

. 2020 Sep 12;22(9):1020. doi: 10.3390/e22091020

Table 4.

Predictions of three classifiers for five tweets.

${\hat{t}}_{1}$	${\hat{t}}_{2}$	${\hat{t}}_{3}$	${\hat{t}}_{4}$	${\hat{t}}_{5}$
$R_{1} = N O N E$	$R_{2} = N E U$	$R_{3} = P$	$R_{4} = P$	$R_{5} = N$
$a_{1, 1} = P$	$a_{1, 2} = N O N E$	$a_{1, 3} = P$	$a_{1, 4} = N$	$a_{1, 5} = N$
$a_{2, 1} = N$	$a_{2, 2} = P$	$a_{2, 3} = P$	$a_{2, 4} = N O N E$	$a_{2, 5} = N$
$a_{3, 1} = N O N E$	$a_{3, 2} = P$	$a_{3, 3} = N$	$a_{3, 4} = N E U$	$a_{3, 5} = N$
$d_{1} = 1$	$d_{2} = 0$	$d_{3} = 1$	$d_{4} = 0$	$d_{5} = 1$