Hybrid Genetic Algorithm Random Forest algorithm (HGARF) for improving the missing value Imputation in Hepatitis Medical Dataset

It has always been difficult to make an exact prediction when there are many missing numbers in the data collection. To solve this problem, the majority of hybrid models either removed the absent cases from the data collection or used a default method to fill in the blanks. In the study described in this article, an improved genetic algorithm metaheuristics-based forecasting model is suggested to forecast the number of confirmed instances of hepatitis. The Random Forest's characteristics are chosen using the improved genetic method, which also improves the efficiency and accuracy of the prediction model overall. On standard medical data sets for Hepatitis from the UCI Repository of Machine Learning, the effectiveness of the suggested model as a predictive categorization system is examined. In addition to accuracy, sensitivity, specificity; and the area under ROC are also computed. The proposed method was contrasted with various current state-of-the-art approaches that were examined in the same setting and using the same datasets. Based on the outcomes of the simulations and the comparative analysis that was done, it was found that the suggested hybrid method outperformed more complex ones with an accuracy of 93.91% and that it could be applied to other prediction models.