Towards Distributed Association Rule Mining Privacy
Mafruz Ashrafi (Monash University, Australia), David Taniar (Monash University, Australia) and Kate Smith (Monash University, Australia)
Copyright: © 2007
With the advancement of storage, retrieval, and network technologies today, the amount of information available to each organization is literally exploding. Although it is widely recognized that the value of data as an organizational asset often becomes a liability because of the cost to acquire and manage those data is far more than the value that is derived from it. Thus, the success of modern organizations not only relies on their capability to acquire and manage their data but their efficiency to derive useful actionable knowledge from it. To explore and analyze large data repositories and discover useful actionable knowledge from them, modern organizations have used a technique known as data mining, which analyzes voluminous digital data and discovers hidden but useful patterns from such massive digital data. However, discovery of hidden patterns has statistical meaning and may often disclose some sensitive information. As a result, privacy becomes one of the prime concerns in the data-mining research community. Since distributed data mining discovers rules by combining local models from various distributed sites, breaching data privacy happens more often than it does in centralized environments.