Publication

Abstract : Heart failure is a common event caused by Cardiovascular diseases which causes major death count and several diagnosis methods were also involved. But still the failure rate prediction is lacking because of medical examination as well as tools used. This paper explores the meticulousness of a machine learning and artificial intelligence based automatic prediction model, which is built by IBM services for heart failure rate prediction where the dataset is trained and a model is built. The auto AI instance is created in the IBM Watson Studio and machine learning services are linked with it. The auto AI service determines the best algorithm as the Gradient Boost algorithm for the given dataset here and automatically classifies it as a binary classification problem with values as Y/N for heart failure. Several algorithms can be chosen and deployed. The NodeRED service is used to deploy the model as a final application. The accuracy along with precision and recall measures and metrics were chosen automatically by the system as best ones. The infographics of the results determines that several other algorithms can also be merged and executed one. Also it is evident from the results, that with a minimum span of time, the application is automatically modeled and deployed for the major threatening disease.