Database Reference
In-Depth Information
Chapter 9
Border Revision for Knowledge Hiding
In this chapter, we highlight the process of border revision, which plays a significant
role on both border-based and exact approaches for association rule hiding. Simply
stated, the process of border revision captures those itemsets of the original database
which need to remain frequent and those that need to become infrequent in the
sanitized version of the database, in order to allow furnishing an optimal hiding
solution. After presenting the theory behind border revision, we introduce a set of
algorithms (adopted from [27]) that can be applied for the efficient computation of
both the original and the revised borders in a transactional database, when given a
minimum support threshold.
Table 9.1: An original database D
O
.
a
b
c
d
e
f
1
1
0
0
0
1
1
1
1
1
0
0
1
0
1
0
0
1
1
0
0
0
0
0
0
1
0
0
1
0
1
1
1
1
1
0
0
0
0
1
0
0
1
1
1
0
1
0
0
1
1
0
0
0
1
0
1
1
1
0
Borders [46] allow for a condense representation of the frequent itemsets in a
database, effectively identifying those key itemsets in the lattice which separate the
frequent patterns from their infrequent counterparts. The process of border revision
facilitates the minimum harm in the hiding of the sensitive itemsets. In what fol-
lows, we use a simple example to demonstrate how this process works. Consider
database D
O O
using mfreq
= 0:3 leads to the set of frequent itemsets shown in
Table 9.2.
Among these item-
of Table 9.1. Applying frequent itemset mining in D
