June 2017
Beginner to intermediate
576 pages
15h 22m
English
Content preview from Practical Predictive Analytics
Become an O’Reilly member and get unlimited access to this title plus top books and audiobooks from O’Reilly and nearly 200 top publishers, thousands of courses curated by job role, 150+ live events each month,
Start your free trial
Variable influence plot
Now we can run our variable influence plot. The following plot indicates high cholesterol and family history as the most important variables:
require(randomForest) fit <- randomForest(factor(heartdisease)~., data=heart,ntree=1000) (VI_F <- importance(fit)) varImpPlot(fit,type=2,main="Random Forest Variable Importance Plot - Heart Disease Simulation")
A downside to this method is that it is treating each variable individually, and not considering any correlation between two variables.