Data Science Desktop Survival Guide by Graham Williams
Desktop Survival Project Home Preface Data Science Introducing R R Constructs R Tasks R Strings R Read, Write, and Create Data Template Data Exploration Data Wrangling Data Visualisation Statistics ML Template ML Scenarios ML Activities ML Applications ML Algorithms Cluster Analysis Decision Trees Computer Vision Graph Data Privacy Literate Data Science Coding with Style Resources Bibliography Index

CLICK HERE TO VISIT THE UPDATED SURVIVAL GUIDE

Decision Tree Performance

Here we plot the performance of the decision tree, showing a risk chart. The areas under the recall and risk curves are also reported.

predict_te <- predict(model, ds[te, vars], type="prob")[,2]
riskchart(predict_te, actual_te, risk_te)

An error matrix shows, clockwise from the top left, the percentages of true negatives, false positives, true positives, and false negatives.

predict_te <- predict(model, ds[te, vars], type="class")
sum(actual_te != predict_te)/length(predict_te) # Overall error rate

## [1] 0.1616943

actual_te %>%
  table(predict_te, dnn=c("Actual", "Predicted")) %>%
  '*'(100/(length(predict_te))) %>%
  round()

##       Predicted
## Actual No Yes
##    No  77   3
##    Yes 13   7