University of California, Irvine (UCI) created a dataset of 299 patients who have been diagnosed with heart failure. There were 13 clinical features that were recorded: age, anemia, high blood pressure, creatinine phosphokinase, diabetes, ejection fraction, platelets, sex, serum creatinine, serum sodium, smoking, time, and death event. The data analysis, inserted below, investigates the data with two different methods: a decision tree and logistic regression. A decision tree, which is an algorithm, breaks the data into different categories based off of the variables while logistic regression is a data model that estimates the effect each variable has on the outcome. The data was also represented graphically using box plots. Since heart failure affects so many people, it is important to understand what increases a person’s chances of dying from heart failure once they have been diagnosed with it. Dive into the attached document below to analyze what variables are associated with individuals dying after their diagnosis of heart failure.
