BagMOOV: A novel ensemble for heart disease prediction bootstrap aggregation with multi-objective optimized voting

摘要: Conventional clinical decision support systems are based on individual classifiers or simple combination of these which tend to show moderate performance. This research paper presents a novel classifier ensemble framework enhanced bagging approach with multi-objective weighted voting scheme for prediction and analysis heart disease. The proposed model overcomes the limitations conventional performance by utilizing an five heterogeneous classifiers: Naive Bayes, linear regression, quadratic discriminant analysis, instance learner vector machines. Five different datasets used experimentation, evaluation validation. obtained from publicly available data repositories. Effectiveness is investigated comparison results several classifiers. Prediction assessed ten fold cross validation ANOVA statistics. experimental shows that deals all type attributes achieved high diagnosis accuracy 84.16 %, 93.29 % sensitivity, 96.70 specificity, 82.15 f-measure. f-ratio higher than f-critical p value less 0.05 95 confidence interval indicate extremely statistically significant most datasets.