Optimizing Random Forest using Genetic Algorithm for Heart Disease Classification

Abstract

Heart disease is a leading cause of death worldwide, and the need for effective predictive systems is a major source of the need to treat affected patients. This study aimed to determine how to improve the accuracy of Random Forest in predicting and classifying heart disease. The experiments performed in this study were designed to select the most optimal parameters using an RF optimization technique using GA. The Genetic Algorithm (GA) is used to optimize RF parameters to predict and classify heart disease. Optimization of the Random Forest parameter using a genetic algorithm is carried out by using the Random Forest parameter as input for the initial population in the Genetic Algorithm. The Random Forest parameter undergoes a series of processes from the Genetic Algorithm: Selection, Crossover Rate, and Mutation Rate. The chromosome that has survived the evolution of the Genetic Algorithm is the best population or best parameter Random Forest. The best parameters are stored in the hall of fame module in the DEAP library and used for the classification process in Random Forest. The optimized RF parameters are max_depth, max_features, n_estimator, min_sample_leaf, and min_sample_leaf. The experimental process performed in RF uses the default parameters, random search, and grid search. Overall, the accuracy obtained for each experiment is the default parameter 82.5%, random search 82%, and grid search 83%. The RF+GA performance is 85.83%; this result is affected by the GA parameters are generations, population, crossover, and mutation. This shows that the Genetic Algorithm can be used to optimize the parameters of Random Forest.

Max Depth Impact on Heart Disease Classification: Decision Tree and Random Forest

“…Based on Table 5, the classification of heart disease using the Decision Tree algorithm and Random Forest using the analysis of the influence of the max depth parameter (3,4,5,6,7) produces the best accuracy of 99.29% in the Random Forest algorithm with max depth = 7, and Decision Tree 98.05% with max depth = 7 on split training and testing data 90:10. The higher the max depth value in research [20], [28] states that it can increase accuracy, but other factors such as split data also affect accuracy results, such as in split data 60:40, the accuracy on Decision Tree testing data max depth = 3 produces 82.68% accuracy, but at max depth = 4, the accuracy decreases to 82.19%.…”

“…While research [18], [19] shows in other studies that if the max depth in Random Forest, is properly optimized, it can produce a classification model that outperforms Decision Tree. Research [20] states that the greater the max depth value can increase the accuracy. However, no research specifically examines the effect of max depth on the classification of heart disease using the Decision Tree and Random Forest methods.…”

A genetic algorithm-based feature selection approach for diabetes prediction

“…The result achieved by the model was 95.60%. Togatorop et al [18] proposed stacked generalization GA to predict heart disease. PID and Diabetes 130-US hospital datasets were used to check the validity of the model.…”

“…[13]- [15], [19], [22], [24] 6 LR [15], [22], [26] 3 NB [14], [17], [26] 3 RF [13], [14], [16], [17], [18], [22], [23], [26], [27] 9 ANN…”