Finding minimum confidence threshold to avoid derived rules in association rule mining

Finding minimum confidence threshold to avoid derived rules in association rule mining

Nzar Abdulqader Ali
1 School of Administration and Economy, University  of Sulaiman, New Campus, Sulaimaniyah, Kurdistan Region - Iraq
Email: nzar@mail.com 




Article info

Original: 30 May 2015
Revised: 14 June 2015
Accepted: 25 June 2015
Published online: 
20 Dec. 2015 

Key Words:
Data Mining
Association Rule
Privacy Preserving


Abstract
In this paper, we focus on obtaining an approximate solution of the two types of two-dimensional linear Volterra-Fredhom integral equations of the second kind. Series solution method is Data in data warehouse often contains sensitive information, the concept of Privacy-Preserving has recently been proposed in response to the concerns of preserving sensitive information derived from published rules. A number of privacy preserving data publishing (PPDP) have been proposed. In this paper an algorithm proposed for hiding published rules that leads to disclosure of sensitive information by determining the confidence value of those rules from the raw data before running association rule mining using prior and posterior probabilities of generated rules and pass those confidence values to data miner to take it in his account when determining minimum confidence threshold in association rule mining algorithms .The experimental results show that the run time for deriving sensitive rules is stabile for different confidence values in comparison with other methods running linear programming methods for finding sensitive published rules. The most derived rules from goal rules (the rules derived from sensitive rules with minimum confidence value) located between 0.5 and 0.8 and these range of confidence values are critical values for data miner, finally experimental results shows that with support values %40,%58, and  %63 still  there is amount of derived published rules appears, and these results means that even with large minimum support threshold still derived published rules appears in association rule algorithms. 


Ċ
Kewan Omer,
Dec 22, 2015, 3:37 PM