Balanced Approach for Hiding Sensitive Association Rules in Data Sharing Environment

Balanced Approach for Hiding Sensitive Association Rules in Data Sharing Environment

Janakiramaiah Bonam (DVR & Dr HS MIC College of Technology, Kanchikacherla, India) and Ramamohan Reddy (Sri Venkateswara University, Tirupati, India)
Copyright: © 2014 |Pages: 24
DOI: 10.4018/IJISP.2014070103
OnDemand PDF Download:
$30.00
List Price: $37.50

Abstract

Privacy preserving association rule mining protects the sensitive association rules specified by the owner of the data by sanitizing the original database so that the sensitive rules are hidden. In this paper, the authors study a problem of hiding sensitive association rules by carefully modifying the transactions in the database. The algorithm BHPSP calculates the impact factor of items in the sensitive association rules. Then it selects a rule which contains an item with minimum impact factor. The algorithm alters the transactions of the database to hide the sensitive association rule by reducing the loss of other non-sensitive association rules. The quality of a database can be well maintained by greedily selecting the alterations in the database with negligible side effects. The BHPSP algorithm is experimentally compared with a HCSRIL algorithm with respect to the performance measures misses cost and difference between original and sanitized databases. Experimental results are also mentioned demonstrating the effectiveness of the proposed approach.
Article Preview

Introduction

In recent years, significant advances in data collection and data-storage technologies have provided the means for inexpensive storage of the huge volume of transactional data in data warehouses that reside in governments, corporations and non-profit organizations. Apart from the benefit of using this data for transaction processing, mining of these databases with the existing data mining tools can reveal hidden, novel, and valuable knowledge. The extracted knowledge patterns can offer insight to the data holders as well as be valuable in important jobs, such as decision making and strategic preparation. Moreover, companies are often willing to cooperate with other entities who conduct similar business, towards the mutual benefit of their businesses. Successful collaboration can bring products to market more rapidly, reduce fabrication and logistics costs and increases sales (Corsten & Danie, 2005). Significant knowledge patterns can be derived and shared among the partners through mutual mining of their databases. Consider the business world as an example to discover the benefits of data sharing. As a case, we present the Wal-Mart and P&G’s story (Grean & Shaw, 2002). Wal-Mart is a major retailer in the US and P&G is an international manufacturer. The two companies agreed in 1988 to share information and knowledge across their mutual supply chains in order to organize their activities in a more serious manner. The partnership of the two companies resulted in the betterment of their business relationship and reduced the associated prices. This resulted in a sharp increase in the joint sales of the two companies. By 1990, the two companies had achieved to significantly improve their everyday business towards their mutual profitability. The vast amount of customer data owned by Wal-Mart is shared to establish more business collaborations with external partners in less time. Wal-Mart continued to share portions of its consumer sales data with larger Market research companies. Strategic sales information about Wal-Mart was used in the research and industry-wide reports were broadly spread, even to the company’s business competitors. The business managers of Wal-Mart soon identified that the dissemination of customer sales data to their partners lead to leakage of their business strategic knowledge to its business competitors, proving that this business strategy of sharing the data as a bad idea (Heun, 2001).

It becomes apparent from the previous discussion that there exists an expanded circle of application scenarios in which collected data or knowledge patterns extracted from the data have to be shared with other entities, which are not trusted, to serve owner specific or organization specific purposes. The sharing of data and/or knowledge may come at a cost of privacy. If the data was related to business or organization's information, then disclosure of this data or any knowledge extracted from the data may potentially reveal sensitive trade secrets, whose knowledge can provide a significant advantage to business competitors and thus can cause the data owner to lose business over his or her peers. Intelligent analysis of data through inference based attacks, may uncover the sensitive patterns of the persons who are not trusted. This can be primarily achieved by utilizing external and publicly available sources of information in conjunction with the published data or knowledge to re-identify individuals or reveal hidden knowledge patterns (Farkas & Jajodia,2002; Morgenstern, 1988). Therefore, compliance with privacy regulations requires the incorporation of sophisticated and refined privacy preserving methodologies.

Complete Article List

Search this Journal:
Reset
Open Access Articles: Forthcoming
Volume 12: 4 Issues (2018): 1 Released, 3 Forthcoming
Volume 11: 4 Issues (2017)
Volume 10: 4 Issues (2016)
Volume 9: 4 Issues (2015)
Volume 8: 4 Issues (2014)
Volume 7: 4 Issues (2013)
Volume 6: 4 Issues (2012)
Volume 5: 4 Issues (2011)
Volume 4: 4 Issues (2010)
Volume 3: 4 Issues (2009)
Volume 2: 4 Issues (2008)
Volume 1: 4 Issues (2007)
View Complete Journal Contents Listing