Generate, Visualise, and Evaluate Fast-and-Frugal Decision Trees
Add an FFT definition to tree definitions
Add nodes to an FFT definition
Add decision statistics to data (based on frequency counts of a 2x2 cl...
Compute classification statistics for binary prediction and criterion ...
Fit and predict competing classification algorithms
Describe data
Drop a node from an FFT definition
Edit nodes in an FFT definition
Clean factor variables in prediction data
Open the FFTrees package guide
Main function to create and apply fast-and-frugal trees (FFTs)
Apply an FFT to data and generate accuracy statistics
Create an object of class FFTrees
Calculate thresholds that optimize some statistic (goal) for cues in d...
Create FFT definitions
Describe a fast-and-frugal tree (FFT) in words
Fit competitive algorithms
Grow fast-and-frugal trees (FFTs) using the fan
algorithms
Rank FFTs by current goal
Perform a grid search over factor and return accuracy statistics for a...
Perform a grid search over thresholds and return accuracy statistics f...
Convert a verbal description of an FFT into an FFTrees
object
Flip exits in an FFT definition
Select the best tree (from current set of FFTs)
Get exit type (from a vector x
of FFT exit descriptions)
Get FFT definitions (from an FFTrees
object x
)
Cue costs for the heartdisease data
Heart disease testing data
Heart disease training data
Provide a verbal description of an FFT
Plot an FFTrees
object
Predict classification outcomes or probabilities from data
Print basic information of fast-and-frugal trees (FFTs)
Read an FFT definition from tree definitions
Reorder nodes in an FFT definition
Select nodes from an FFT definition
Visualize cue accuracies (as points in ROC space)
Summarize an FFTrees
object
Write an FFT definition to tree definitions
Create, visualize, and test fast-and-frugal decision trees (FFTs) using the algorithms and methods described by Phillips, Neth, Woike & Gaissmaier (2017), <doi:10.1017/S1930297500006239>. FFTs are simple and transparent decision trees for solving binary classification problems. FFTs can be preferable to more complex algorithms because they require very little information, are easy to understand and communicate, and are robust against overfitting.
Useful links