Database Reference
In-Depth Information
Algorithm 10.1 The Border Based Approach of Sun & Yu [66, 67].
1: function BBA(Original database D O , frequent itemsets F D O , sensitive itemsets S, minimum
support threshold msup)
2: D 0 D O
3: Compute S min and Bd +
4:
Sort itemsets in S min in decreasing order of length and increasing order of support
5:
for each I 2S min do
Compute Bd + j I and w(I 2Bd + j I )
6:
7:
Initialize the set Cj I of hiding candidates for itemset I
for i = 0; i < sup(I;D 0 )msup+1; i++ do
8:
. candidate selection
9: Find hiding candidate c = (T o ; i o ) with minimal impact in C
10: C Cf(T; i)jT = T o g
11: Update w(I 2Bd + j I )
12: end for
13: Update database D 0
14: end for
15: Return: sanitized database D D 0
16: end function
10.3 Algorithm
Algorithm 10.1 summarizes the border based approach for the hiding of the sensitive
itemsets. At each step the algorithm selects the hiding candidate that minimally
affects the itemsets of the revised positive border and applies the corresponding
item deletion. The details of the candidate selection process (see the inner for-loop)
are beyond the scope of this chapter. The interested reader can refer to [67].
 
Search WWH ::




Custom Search