Cancer Classification Utilizing Voting Classifier with Ensemble Feature Selection Method and Transcriptomic Data

. 2023 Sep 14;14(9):1802. doi: 10.3390/genes14091802

Algorithm 2 Majority voting ensemble classifier

Input:
- L: Labeled tuples for training of each class C with selected features.
- D: Set of tuples for evaluation.
- T: Set of tuples for testing.
- N: (

M L_{1}

M L_{2}

M L_{3}

, …,

M L_{N}

) Set of classifier ML algorithm.

1:
for each ML classifier $M L_{i} \in {1, \dots, N}$ do
2:
Train $M L_{i}$ using labeled data L
3:
end for
4:
for each ML classifier $M L_{i} \in {1, \dots, N}$ do
5:
Test and evaluate $M L_{i}$ using testing data D
6:
end for
7:
for each ML classifier $M L_{i} \in {1, \dots, N}$ do
8:
Assign weight of $M L_{i}$ based on the accuracy of $M L_{i}$
9:
end for
10:
for each each test data $t_{i} \in {T}$ do
11:
for each ML classifier $M L_{i} \in {1, \dots, N}$ do
12:
Find the predicted class of $t_{i}$
13:
end for
14:
Aggregate vote to ensemble based on the results
15:
end for