An Efficient Swarm Optimized Based Machine-Learning Technique for Improving the Prediction of Chronic Diseases

Heart Disease is considered as one of the deadliest chronic diseaseChronic diseases across the globe. In spite of the advancement in the healthcareHealthcare technology, the mortality rate due to the cardiovascular disease increasing every day. The heart disease will claim at least 18 million lives each year. The manual based treatment to diseased patient by physician is one of the solution to reduce the mortality rate. But one of the drawback of this approach is, it will consume more time as sign of the disease will not be known in advance. In order to overcome with these problems in the proposed paper machine learningMachine learning based algorithmic model has been suggested to early predict the chronic diseaseChronic diseases. In the proposed paper machine learning algorithmsMachine learning algorithms such as Random ForestRandom Forest (RF), Decision TreeDecision Tree (DT), SVMSupport Vector Machine (SVM), and Naive BayesNaive Bayes classifier and stacked classifier is applied on the heart disease dataset to early predict the disease. The performance of the machine learning algorithmsMachine learning algorithms are further optimized using proposed Hybrid DragonflyDragonfly-Genetic Algorithm. Further the performance of the suggested model is subsequently validated by using the various performance metrics. The experimental results shows that the combination of Hybrid DragonflyDragonfly-Genetic Algorithm and stacked classifier attained higher accuracy compared to the existing algorithms.