Research Article

Efficiently Hiding Sensitive Itemsets with Transaction Deletion Based on Genetic Algorithms

Algorithm 1

cpGA2DT algorithm.
Input:   , , ,   .
Output: A sanitized database .
Termination condition: The fitness  := 0 or the number of generation  := .
(1) set .
(2) scan to get and respectively by and .
(3) for (, ; , ) do
   if () then
    project from to form .
   end if
  end for
// initialize the probability vector for each transaction in .
(4) for (, ) do
  := 0.5.
  end for
// generate two individuals with m genes from by .
(5) := or 0, , 1 ≤ .
(6) := or 0, , 1 ≤ .
// compete   and   .
(7) winner, loser:= compete(, ) by fitness.
// update the probability vector towards to the better chromosome.
(8) for (, ) do
  := / for the of winner.
  := / for the of loser.
  end for
(9) if   terminated condition is not satisfied then
 perform Steps 5 to 8.
  else
 terminate.
  end if