| Title: | Supervised Classification Learning and Prediction using Patient Rule Induction Method (PRIM) |
| Version: | 2.0.0 |
| Date: | 2016-10-01 |
| Description: | The Patient Rule Induction Method (PRIM) is typically used for "bump hunting" data mining to identify regions with abnormally high concentrations of data with large or small values. This package extends this methodology so that it can be applied to binary classification problems and used for prediction. |
| Depends: | R (≥ 3.1.1), stats, prim (≥ 1.0.16) |
| Suggests: | kernlab, testthat |
| License: | GPL-3 |
| URL: | https://github.com/dashaub/supervisedPRIM |
| BugReports: | https://github.com/dashaub/supervisedPRIM/issues |
| LazyData: | true |
| RoxygenNote: | 5.0.1 |
| ByteCompile: | true |
| NeedsCompilation: | no |
| Packaged: | 2016-10-01 02:46:22 UTC; david |
| Author: | David Shaub [aut, cre] |
| Maintainer: | David Shaub <davidshaub@gmx.com> |
| Repository: | CRAN |
| Date/Publication: | 2016-10-01 14:39:25 |
Model Predictions
Description
Perform prediction on a trained supervisedPRIM model. Output
to either predicted class or positive class probability is supported.
Usage
## S3 method for class 'supervisedPRIM'
predict(object, newdata, classProb = FALSE, ...)
Arguments
object |
A trained model of class |
newdata |
The new data on which to create predictions |
classProb |
Should the function return the estimated class |
... |
additional arguments (ignored) probabilities instead of the predicted class? |
Author(s)
David Shaub
Examples
# Train a model to determine if a flower is setosa
data(iris)
yData <- factor(ifelse(iris$Species == "setosa", "setosa", "other"), levels = c("setosa", "other"))
xData <- iris
xData$Species <- NULL
primModel <- supervisedPRIM(x = xData, y = yData)
# Predict on the original dataset
predictions <- predict(primModel, newdata = xData)
Fit PRIM model to a labeled dataset
Description
perform supervised classification using Patient Rule Induction Method (PRIM)
Usage
supervisedPRIM(x, y, peel.alpha = 0.05, paste.alpha = 0.01,
mass.min = 0.05, threshold.type = 1, ...)
Arguments
x |
matrix of data values |
y |
binary vector of 0/1 response values |
peel.alpha |
peeling quantile tuning parameter |
paste.alpha |
pasting quantile tuning parameter |
mass.min |
minimum mass tuning parameter |
threshold.type |
threshold direction indicator: 1 = ">= threshold", -1 = "<= threshold" |
... |
additional arguments to pass to |
Details
Fit
Value
an object of class supervisedPRIM. See additional details
in prim.box
Author(s)
David Shaub
Examples
# Train a model to determine if a flower is setosa
data(iris)
yData <- factor(ifelse(iris$Species == "setosa", "setosa", "other"), levels = c("setosa", "other"))
xData <- iris
xData$Species <- NULL
primModel <- supervisedPRIM(x = xData, y = yData)