Benchmarking Analysis of the Accuracy of Classification Methods Related to Entropy

. 2021 Jul 1;23(7):850. doi: 10.3390/e23070850

Algorithm 2 Pseudo-code of the algorithm to predict with a CRS model.

1:
INPUT:
2:
s: test row with $C - 1$ antecedent attributes
3:
$C R S$ : Classification Rule Set
4:
USE: RSS: Rule subset
5:
START ALGORITHM
6:
for $c = C - 1$ to 1 do
7:
RSS ← {Ri ∈ CRS / c attributes of s are equal to c attributes of Ri}
8:
if RSS $\neq ⌀$ then
9:
R ← R1 {/* R1 is the first rule of RSS */}
10:
for $j = 2$ to $| R S S |$ do
11:
if R.supp < Rj.supp then
12:
R ← Rj
13:
else if R.supp = Rj.supp and R.conf < Rj.conf then
14:
R ← Rj
15:
end if
16:
return R.consequent
17:
end for
18:
end if
19:
end for
20:
return The resulting predicted class for row s (the consequent of a rule of CRS) {/*OUTPUT*/}
21:
END ALGORITHM