3.2.1. Initial Population

The dataset utilized in this work consisted of 78 features, which was a relatively high number and could affect the quality of the output. Therefore, among these 78, the GA chose 25 features randomly to start the initial population. Knowing that the GA was tested with different numbers of features, from 10 to 25, the results were close, and the training and testing times of the data were adequate. After experimenting with 30 to 40 of the selected features, a decrease in the performance of the system was observed, as the accuracy percentage decreased and the times for training and testing data increased.
