Accuracy Comparison of Random Tree and Random Forest and Kstar Classification Algorithm for Angiographic Disease Status Prediction
Emergency service provision and intensive care provision of cardiac patients are some areas where automation can allieviate the stress on healthcare centres. Data from personal monitoring devices, android applications to IOT can be used for continuous and ubiquitous monitoring of intensive care patients, the obtained analysis can be remotely observed my medical personnel. The efficient analysis, diagnosis and decision making in the healthcare is paramount in case of intensive care patients. One of the most efficient manner of implementation is by integration of data mining techniques in the present legacy systems of healthcare sectors. The early detection and prediction of emergency service requirement for intensive care patients is one such area where automation can save lives. The proposed work here uses data gathered from patients to predict and generate the patterns or rules which can classify the patient’s tendency to have angiographic disease. This paper assess the classification ability of Random tree, Random Forest and Kstar algorithm accuracy in predicting angiographic disease status.
Keywords - Data Mining, Health Informatics, Classifiers, Healthcare, Star, Random Tree, Random Forest.