Computes the specificity of class prediction

Description

Computes the specificity of class prediction.

Usage

## S3 method for class 'BigBang':
specificityClass(o, cm, ...)

Arguments

Details

Specificity is the probability that a sample of class different to X will NOT be predicted as class X. High specificity avoids false positives. Specificity = TN / (TN + FP) TN - True Negatives: For class A, TN = Pbb + Pbc + Pbx + Pcb + Pcc + Pcx FP - False Positives: For class A, FP = Pba + Pca Confusion Matrix: [ Predicted Class ] ClassA ClassB ClassC "misclass" ClassA Paa Pab Pac Pax ClassB Pba Pbb Pbc Pbx ClassC Pca Pcb Pcc Pcx

Value

A vector with the specificity of prediction for every class.

Author(s)

Victor Trevino. Francesco Falciani Group. University of Birmingham, U.K. http://www.bip.bham.ac.uk/bioinf

References

Goldberg, David E. 1989 Genetic Algorithms in Search, Optimization and Machine Learning. Addison-Wesley Pub. Co. ISBN: 0201157675

See Also

For more information see BigBang. *classPredictionMatrix(), *confusionMatrix().

Examples

   #bb is a BigBang object
   cpm <- classPredictionMatrix(bb)
   cpm
   cm <- confusionMatrix(bb)
   cm
   #equivalent and quicker because classPredictionMatrix is provided
   cm <- confusionMatrix(bb, cpm)
   cm
 
   specificityClass(bb, cm)
   specificityClass(bb, cpm)
   specificityClass(bb)
   # all are equivalent
   sensitivityClass(bb, cpm)
   sensitivityClass(bb, cm)
   sensitivityClass(bb)
   # all are equivalent