An Efficient Algorithm for the Detection of Outliers in Mislabeled Omics Data

Algorithm 1

Description of C-step algorithm.

While (continueCstep)

do

Penalized logisitic regression is applied on the current subset , and get

   For , of every observation is got and observations are sorted according to their deviances.

      ,

   The h observations with smallest negative log-likelihood function are retained to form a subset .

end