Exploiting Machine Learning to Detect Malicious Nodes in Intelligent Sensor-Based Systems Using Blockchain

Algorithm 3

Detection of malicious nodes with classification by GA-DT.

1 load dataset;

2 split the data 70% = testing and 30% = training;

3 initialize hyperparameters (Maximum Depth, Minimum Sample Leaf, Minimum Sample Split, Criterion);

4 GA:Maximum Depth, Minimum Sample Leaf, Minimum Sample Split, Criterion;

5 initialize population

6 (generation,populationsize,offspringsize);

7 perform tuning of hyperparameters;

8 perform selection;

9 perform crossover;

10 perform mutation;

⊳Decision Tree is DT

11 train DT on selected parameters;

12 DT saves the model;

13 DT tests the model;

14 DT loads the model;