ClustEval clustering evaluation framework
The FDR relates the number of false positives to the total number of positively predicted elements. This estimates a likelihood of an element being negative, if it is predicted positive. This measure is based on the pairwise approach to calculate TP,TN,FP and FN. $FDR = \frac{FP}{FP+TP}$ In the binary classification background we have two classes that we want to distinguish: positive and negative. In this scenario there are four possible outcomes: