Skip to content

Releases: Byron1001/strokePrediction

Shiny Framework release

23 May 04:11
Compare
Choose a tag to compare

After processing with the data pre-processing and up until processing with modeling the outcome has been determined. When passing the analysis, the most suitable and accurate model in our analysis will be RANDOM FOREST after comparing with a total of 6 models which include LightGBM, KNN, decision tree, SVM, random forest, and logistic regression. We have identified that age, marital status, heart disease, BMI, and average glucose level are the most important indicators that determine whether one has a stroke. By implementing the shiny framework, our work enables people to simply enter the data that are commonly known to predict their stroke likelihood.