PRIVACY-PRESERVING COLLABORATIVE DATA MINING

Data mining is a process to extract useful knowledge from large amounts of data. To conduct data mining, we often need to collect data from various resources. However, the data are sometimes distributed among and owned by different parties. Privacy concerns may prevent the parties from directly sharing the actual values of data and some types of information about the data. How multiple parties can collaboratively conduct data mining without breaching data privacy presents a grand challenge. Theoretical results from the area of secure multi-party computation show that one may provide secure protocols for any multi-party computation with honest majority. However, the general methods are far from efficient and practical for computing complex functions on inputs consisting of large sets of data. Therefore, to efficiently tackle the problem, formulated as Privacy-Preserving Collaborative Data Mining (PPDM), we need to develop privacy-conscious solutions with adequate efficiency. Our goal is to provide efficient solutions to the problem of data sharing among multiple parties involved in a data mining task, without disclosing the data between the parties. We have developed various privacy-oriented protocols for multiple parties to conduct the desired data mining tasks. We provide efficient solutions to obtain accurate data mining results and minimize private data disclosure. The solutions are distributed, i.e., there is no centralized, trusted party having access to all the data. Instead, we develop secure protocols to exchange the data while preserving the data privacy. In this talk, I will introduce the challenges of PPDM, present solutions, and sketch future directions for this research.