|International Journal of Applied Information Systems|
|Foundation of Computer Science (FCS), NY, USA|
|Volume 12 - Number 8|
|Year of Publication: 2017|
|Authors: Kshitij Pathak, Sanjay Silakari, Narendra S. Chaudhari|
Kshitij Pathak, Sanjay Silakari, Narendra S. Chaudhari . Privacy Preserving Informative Association Rule Mining. International Journal of Applied Information Systems. 12, 8 ( Nov 2017), 1-7. DOI=10.5120/ijais2017451717
Privacy preserving data mining has two major directions: one is the protection of private data, i.e., data hiding in the database whereas another one is the protection of sensitive rule (Knowledge) contained in data known as knowledge hiding in the database. This research work focuses on protection of sensitive association rule. Corporation individual & other may get mutual benefit by sharing their data, but at the same time, they would like to be sure that their sensitive data remains private or not disclosed, i.e., hiding sensitive association rules. Approaches need to be given sensitive association rule in advance to hide them, i.e., mining is repaired. However, for some application pre-process of these sensitive association rules is combined with hiding process when predictive items are given, i.e., hiding informative association rule set. In this work, we propose two algorithms ISLFASTPREDICTIVE, DSRFASTPREDICTIVE to hide informative association rule with n-items. Earlier work hided 2-item association rules. Algorithms proposed in the paper execute faster than ISL & DSR algorithms prepared earlier as well as a side effect have been reduced. ISLFASTPREDICTIVE and DSRFASTPREDICTIVE algorithms work better as database scans are reduced since transaction list of elements is used in algorithms, i.e., a list of the transaction which supports itemsets and selection of transactions are done on the basis of presence of frequent itemsets.