Go to TogaWare.com Home Page. Data Science Desktop Survival Guide
by Graham Williams
Duck Duck Go

C5.0 Rules 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.

predicted <- predict(model, ds[te, vars], type="prob")[,2]
riskchart(predicted, actual, risks)

## Error in na.omit(ac): object 'actual' not found

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

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

## Warning: Incompatible methods ("Ops.data.frame", "Ops.factor") for "!="

## Error in ds[te, target] != predicted: comparison of these types is not implemented

round(100*table(ds[te, target], predicted, dnn=c("Actual", "Predicted"))/length(predicted))

## Warning in Ops.factor(xi, xj): '>' not meaningful for factors

## Error in if (xi > xj) 1L else -1L: missing value where TRUE/FALSE needed

Support further development by purchasing the PDF version of the book.
Other online resources include the GNU/Linux Desktop Survival Guide.
Books available on Amazon include Data Mining with Rattle and Essentials of Data Science.
Popular open source software includes rattle and wajig.
Hosted by Togaware, a pioneer of free and open source software since 1984.
Copyright © 2000-2020 Togaware Pty Ltd. . Creative Commons ShareAlike V4.