GUI based Prediction of Heart Stroke Stages by finding the accuracy using Machine Learning algorithm

Abstract

Many predictive techniques are used and applied in the medical domain such as predicting occurrence, evaluating outcome of diseases and assisting clinicians to recommend treatment of diseases. Standard predictive models or methods, on the other hand, are incapable of simulating the complexities of feature representation in medical problem domains, and therefore are ineffective in capturing the underlying information. To address this problem, machine learning algorithms are used to apply predictive computational techniques for heart stroke on a given hospital dataset. Atrial fibrillation is a significant risk factor for cardiac attack in patients, and it shares many of the same factors that predict stroke. When a dataset is analysed using a controlled machine learning algorithm, variables such as variable recognition, univariate analysis, bivariate and multivariate analysis, missed value therapies, mathematical methods, and so on are all recorded. The aim of the predictive analytics model is to recognise the various stages of heart stroke in patients. Discuss the output of the provided hospital dataset, as well as the evaluation of the classification study and the uncertainty matrix. To compare supervised classification machine learning algorithms and suggest a machine learning-based method for reliably predicting heart stroke using given characteristics. Furthermore, compare and discuss the performance of different machine learning algorithms from the given healthcare department dataset with evaluation classification report, define the confusion matrix, and categorise data from priority, and the result depicts that the effectiveness of a graphical user interface based proposed machine learning algorithm technique can be compared with best accuracy with precision and F1 Score