USC Predicting the 10 Year Risk of Future Coronary Heart Disease Project – Description
The dataset provided information on over 4,000 patients and included 15 attributes, each representing a potential risk factor for CHD. These attributes included demographic, behavioral, and medical risk factors.Task : “To predict the 10-year risk of future coronary heart disease (CHD) in patients” Use logistic regression and Random forest Use binning and provide examplesWhich features are really important Use Upsampling and downsampling for unbalanced data . And which is the best method and why?Provide cross-validation and performance evaluation metrics specific to imbalanced datasets, to ensure the model’s effectiveness and generalizability.
The post USC Predicting the 10 Year Risk of Future Coronary Heart Disease Project first appeared on .