secure mining of association rules in horizontally distributed databases

Secure Mining of Association Rules in Horizontally Distributed Databases
ABSTRACT
We propose a protocol for secure mining of association rules in horizontally distributed
databases. The current leading protocol is that of Kantarcioglu and Clifton [18]. Our protocol,
like theirs, is based on the Fast Distributed Mining (FDM) algorithm of Cheung et al. [8], which
is an unsecured distributed version of the Apriori algorithm. The main ingredients in our protocol
are two novel secure multi-party algorithms — one that computes the union of private subsets
that each of the interacting players hold, and another that tests the inclusion of an element held
by one player in a subset held by another. Our protocol offers enhanced privacy with respect to
the protocol in [18]. In addition, it is simpler and is significantly more efficient in terms of
communication rounds, communication cost and computational cost.
Existing System
That goal defines a problem of secure multi-party computation. In such problems, there are M
players that hold private inputs, x1, . . . , xM, and they wish to securely compute y = f(x1, . . . ,
xM) for some public function f. If there existed a trusted third party, the players could surrender
to him their inputs and he would perform the function evaluation and send to them the resulting
output. In the absence of such a trusted third party, it is needed to devise a protocol that the
players can run on their own in order to arrive at the required output y. Such a protocol is
considered perfectly secure if no player can learn from his view of the protocol more than what
he would have learnt in the idealized setting where the computation is carried out by a trusted
third party. Yao [32] was the first to propose a generic solution for this problem in the case of
two players. Other generic solutions, for the multi-party case, were later proposed in
Proposed System
Herein we propose an alternative protocol for the secure computation of the union of private
subsets. The proposed protocol improves upon that in [18] in terms of simplicity and efficiency
as well as privacy. In particular, our protocol does not depend on commutative encryption and
oblivious transfer (what simplifies it significantly and contributes towards much reduced
communication and computational costs). While our solution is still not perfectly secure, it leaks
excess information only to a small number (three) of possible coalitions, unlike the protocol of
that discloses information also to some single players. In addition, we claim that the excess
Contact: 9533694296, 9703109334
Email id: academicliveprojects@gmail.com, www.logicsystems.org.in

information that our protocol may leak is less sensitive than the excess information leaked by the
protocol of.
The protocol that we propose here computes a parameterized family of functions, which we call
threshold functions, in which the two extreme cases correspond to the problems of computing the
union and intersection of private subsets. Those are in fact general-purpose protocols that can be
used in other contexts as well. Another problem of secure multiparty computation that we solve
here as part of our discussion is the set inclusion problem; namely, the problem where Alice
holds a private subset of some ground set, and Bob holds an element in the ground set, and they
wish to determine whether Bob’s element is within Alice’s subset, without revealing to either of
them information about the other party’s input beyond the above described inclusion.
IMPLEMENTATION
Implementation is the stage of the project when the theoretical design is turned out into a
working system. Thus it can be considered to be the most critical stage in achieving a successful
new system and in giving the user, confidence that the new system will work and be effective.
The implementation stage involves careful planning, investigation of the existing system and it’s
constraints on implementation, designing of methods to achieve changeover and evaluation of
changeover methods.
Modules:
1. Privacy Preserving Data Mining:
Previous work in privacy preserving data mining has considered two related settings. One, in
which the data owner and the data miner are two different entities, and another, in which the data
is distributed among several parties who aim to jointly perform data mining on the unified corpus
of data that they hold.
In the first setting, the goal is to protect the data records from the data miner. Hence, the data
owner aims at anonym zing the data prior to its release. The main approach in this context is to
apply data perturbation. The idea is that. Computation and communication costs versus the
number of transactions N the perturbed data can be used to infer general trends in the data,
without revealing original record information.
In the second setting, the goal is to perform data mining while protecting the data records of each
of the data owners from the other data owners. This is a problem of secure multiparty
computation. The usual approach here is cryptographic rather than probabilistic.
Contact: 9533694296, 9703109334

Lind ell and Pink as showed how to securely build an ID3 decision tree when the training set is
distributed horizontally. Lin et al. discussed secure clustering using the EM algorithm over
horizontally distributed data. The problem of distributed association rule mining was studied in
in the vertical setting, where each party holds a different set of attributes, and in [18] in the
horizontal setting. Also the work of [26] considered this problem in the horizontal setting, but
they considered large-scale systems in which, on top of the parties that hold the data records
(resources) there are also managers which
2. Distributed Computation:
We compared the performance of two secure implementations of the FDM algorithm
Section In the first implementation (denoted FDM-KC), we executed the unification step using
Protocol UNIFI-KC, where the commutative cipher was 1024-bit RSA in the second
implementation (denoted FDM) we used our Protocol UNIFI, where the keyed-hash function
was HMAC [4]. In both implementations, we implemented Step 5 of the FDM algorithm in the
secure manner that was described in later. We tested the two implementations with respect to
three measures:
1) Total computation time of the complete protocols (FDMKC and FDM) over all players. That
measure includes the Apriori computation time, and the time to identify the globally s-frequent
item sets, as described in later.
2) Total computation time of the unification protocols only (UNIFI-KC and UNIFI) over all
players. 3) Total message size. We ran three experiment sets, where each set tested the
dependence of the above measures on a different parameter: • N — the number of transactions in
the unified database,
3. Frequent Itemsets:
We describe here the solution that was proposed by Kantarcioglu and Clifton. They onsidered
two possible settings. If the required output includes all globally s-frequent item sets, as well as
the sizes of their supports, then the values of Δ(x) can be revealed for all x ∈ Ck s . In such a
case, those values may be computed using a secure summation protocol (e.g. [6]), where the
private addend of Pm is suppm(x) − sNm. The more interesting setting, however, is the one
where the support sizes are not part of the required output. We proceed to discuss it.
Contact: 9533694296, 9703109334

4. Association Rules:
Once the set Fs of all s-frequent itemsets is found, we may proceed to look for all (s, c)-
association rules (rules with support at least sN and confidence at least c), as described in [18].
For X, Y ∈ Fs, where X ∩ Y = ∅, the corresponding association rule X ⇒ Y has confidence at
least c if and only if supp(X ∪ Y )/supp(X) ≥ c, or, equivalently, CX,Y := MΣ m=1 (suppm(X ∪ Y
) − c · suppm(X)) ≥ 0 . (10) Since |CX,Y | ≤ N, then by taking q = 2N+1, the players can verify
inequality (10), in parallel, for all candidate association rules, as described in Section 3. In order
to derive from Fs all (s, c)-association rules in an efficient manner we rely upon the following
straightforward lemma.
Contact: 9533694296, 9703109334

secure mining of association rules in horizontally distributed databases

More Related Content

What's hot (18)

Viewers also liked (7)

Similar to secure mining of association rules in horizontally distributed databases (20)

More from swathi78 (20)

Recently uploaded (20)

secure mining of association rules in horizontally distributed databases