Interactive Pattern Exploration: Securely Mining Distributed Databases
Interactive patterns embedded and stored in multiple related databases can provide valuable insights into the domain of data exploration. Yet, the owners of individual databases may want to protect the privacy of their data while still allowing enough collaboration for the patterns to be discovered. In this paper, we show how data can be accessed securely through the use of data mining algorithms. We also investigate some methods that discover unique data patterns interactively, while still preserving data and user privacy, as much as possible.
KeywordsPrivacy Interaction Security Data ID3 Distributed
The research was supported in part by the National Science Foundation through the REU program (2013–2014) at the University of Cincinnati. We are also thankful to the reviewers for providing useful comments.
- 1.Bhatnagar, R., Srinivasan, S.: Pattern discovery in distributed databases. In: American Association for Artificial Intelligence (1997)Google Scholar
- 2.Chattopadhyay, D.: Distributed decision tree induction using multi-agent based negotiation protocol. Electronic thesis or dissertation. University of Cincinnati, OhioLINK Electronic Theses and Dissertations Center (2014)Google Scholar
- 6.Hsiao, C.J., Cherry, D.K., Rechtsteiner, E.A.: Medical care survey. Centers for Disease Control and Prevention. Centers for Disease Control and Prevention (2013)Google Scholar
- 7.Glinert-Stevens, S.: Microsoft publisher: desktop wizardry. Pc Sources 3(2), 357 (1992)Google Scholar
- 10.Monson, L.: Dr Dobb’s Journal-Software Tools for the Professional Programmer 22(10), pp. 117–120 (1997)Google Scholar
- 11.Olsen, T.C.: Distributed Decision Tree Learning from Multiple Heterogeneous Data Sources (2006)Google Scholar
- 12.Quinlan, J.R.: Induction of decision trees. Mach. Learn. 1(1), 81–106 (1986)Google Scholar