llama (version 0.6)

misclassificationPenalties: Misclassification penalty

Description

Calculates the penalty incurred because of making incorrect decisions, i.e. choosing suboptimal algorithms.

Usage

misclassificationPenalties(data, model)

Arguments

data

the data used to induce the model. The same as given to classify, classifyPairs, cluster or regression.

model

the algorithm selection model. Can be either a model returned by one of the model-building functions or a function that returns predictions such as vbs or the predictor function of a trained model.

Value

• A list of the misclassification penalties.

Details

Compares the performance of the respective chosen algorithm to the performance of the best algorithm for each datum. Returns the absolute difference. This denotes the penalty for choosing a suboptimal algorithm, e.g. the additional time required to solve a problem or reduction in solution quality incurred. The misclassification penalty of the virtual best is always zero.

See Also

parscores, successes

Examples

library(RWeka)

data(satsolvers)
trainTest = cvFolds(satsolvers)

model = classify(classifier=J48, data=trainTest)
sum(misclassificationPenalties(trainTest, model))