Model for Predicting Heart Failure Mortality

Abstract

The aim of this paper is to create a model for predicting heart failure mortality using the decision tree method, and to apply it to a specific set of information using the RapidMiner data analysis platform. Due to the rapid pace of medical development, there is an urgent need to provide recommendations derived from filtering the entire set of medical data. It becomes difficult for patients to make the right decision due to the huge variety of treatments and ways to improve their lifestyle. This necessitates the emergence of predictive methods that aim to offer the patient the most accurate choice according to specific complaints instead of wandering among multiple random choices with varying degrees of significance in reality. Event prediction methods extract the most vital information from a huge data set, “reading” the patient’s problems and suggesting the most appropriate treatment for him/her. Most patients are late in taking the necessary care of their own health, and they realize the importance of prevention at a later stage. If every person could predict what might happen to him/her later in life, he/she would be much more cautious and make more timely efforts for his/her health. The need for decision-making in clinical practice often has important long-term consequences.