Predictive Hybridization Model integrating Modified Genetic Algorithm (MGA) and C4.5

Numerous enhancements of prediction models through hybridization and combining various machine learning to increase the prediction model's performance are still an ongoing research interest in data mining. This study is a modification of GA with a new crossover mating structure called Flip Multi-Sliced Average Crossover (FMSAX) operator with a rank-based selection function integrating the C4.5 algorithm. To measure the accuracy level of the C4.5, the dataset is split into 70:30 for training and testing, respectively. The results showed that the prediction model of the modified GA combined with the C4.5 algorithm outperformed the C4.5 prediction model, the GA having AX and roulette wheel selection function. The results showed that the prediction model obtained an accuracy value of 98.6207%, where its MAE, RMSE, Precision, Recall, and F-Measure values are 0.0066, 0.0738, 0.987, 0.986, and 0.986, respectively.