Heart attack USA heart attack prediction dataset
Richard_Rowe
2025-02-23
A exploration and cleaning of the USA heart attack prediction dataset.
dataset citation listed at end of document.
Summary.
After initial cleaning and analysis the data set was found to contain too much manipulated data. The dataset exhibits an almost perfectly even distribution across several variables, including gender, outcome, education, diet and average age. The relationship between education level and average income is not inline with real world trends. The conclusion is that the dataset is certainly flawed due to manipulation, making the data unreliable for drawing meaningful conclusion about the social economic factors affecting heart attack outcomes.
Click here for full webpage
